Autobiographical Memory and Patterns of Brain
Atrophy in Frontotemporal Lobar Degeneration

Margaret C. McKinnon1,2, Elena I. Nica3, Pheth Sengdy3,
Natasa Kovacevic3, Morris Moscovitch3,4, Morris Freedman3,4,7,8,
Bruce L. Miller4, Sandra E. Black3,4,5, and Brian Levine3,4,6

D
Ö
w
N
l
Ö
A
D
e
D

Abstrakt

& Autobiographical memory paradigms have been increasingly
used to study the behavioral and neuroanatomical correlates of
human remote memory. Although there are numerous func-
tional neuroimaging studies on this topic, relatively few studies
of patient samples exist, with heterogeneity of results owing to
methodological variability. In this study, frontotemporal lobar
degeneration (FTLD), a form of dementia affecting regions cru-
cial to autobiographical memory, was used as a model of auto-
biographical memory loss. We emphasized the separation of
episodic (recollection of specific event, perceptual, and mental
state information) from semantic (factual information unspecific
in time and place) autobiographical memory, derived from a
reliable method for scoring transcribed autobiographical pro-
tocols, the Autobiographical Interview [Levine, B., Svoboda, E.,
Hay, J., Winocur, G., & Moscovitch, M. Aging and autobio-
graphical memory: Dissociating episodic from semantic retrieval.
Psychologie und Altern, 17, 677–689, 2002]. Patients with the
frontotemporal dementia (FTD) and mixed frontotemporal
and semantic dementia (FTD/SD) variants of FTLD were im-
paired at reconstructing episodically rich autobiographical mem-

ories across the lifespan, with FTD/SD patients generating an
excess of generic semantic autobiographical
Information. Pa-
tients with progressive nonfluent aphasia were mildly impaired
for episodic autobiographical memory, but this impairment was
eliminated with the provision of structured cueing, likely reflect-
ing relatively intact medial temporal lobe function, whereas the
same cueing failed to bolster the FTD and FTD/SD patients’
performance relative to that of matched comparison subjects.
The pattern of episodic, but not semantic, autobiographical im-
pairment was enhanced with disease progression on 1- to 2-year
follow-up testing in a subset of patients, supplementing the
cross-sectional evidence for specificity of episodic autobiograph-
ical impairment with longitudinal data. This behavioral pattern
covaried with volume loss in a distributed left-lateralized pos-
terior network centered on the temporal lobe, consistent with
evidence from other patient and functional neuroimaging stud-
ies of autobiographical memory. Frontal lobe volumes, Jedoch,
did not significantly contribute to this network, suggesting that
frontal contributions to autobiographical episodic memory may
be more complex than previously appreciated. &

EINFÜHRUNG

lobar degeneration (FTLD) is a com-
Frontotemporal
mon form of dementia and a leading cause of early-onset
Demenz, along with Alzheimer’s disease (Ratnavalli,
Brayne, Dawson, & Hodges, 2002). Although everyday
memory has been considered relatively spared in FTLD
(Neary et al., 1998), there is evidence that this disease
affects performance on laboratory tests of memory
(Simons, Graham, & Hodges, 2002). An emerging liter-
ature suggests that this deficit extends to the autobio-
graphical domain (Piolino et al., 2003; Nestor, Graham,
Bozeat, Simons, & Hodges, 2002; Hodges & Gurd, 1994).
FTLD is uniquely positioned as a model for testing
hypotheses concerning autobiographical memory. Gen-

1McMaster University, 2St. Joseph’s Healthcare, 3Baycrest Centre,
4Universität von Toronto, 5University of California at San Francisco,
6Sunnybrook and Women’s College Health Sciences Centre,
7Baycrest Centre for Geriatric Care, 8University Health Network

erally speaking, memory impairments in FTLD follow
from degeneration of the frontal and temporal cortex
associated with this disease. The temporal lobes, partic-
ularly the medial temporal regions, are classically asso-
ciated with mnemonic operations. The prefrontal cortex
is involved in higher-order mnemonic retrieval opera-
tionen (Fletcher & Henson, 2001; Stuss & Benson, 1986)
including strategic retrieval and monitoring within auto-
biographical memory (McKinnon, Svoboda, & Levine,
2007; Conway & Pleydell-Pearce, 2000). Genauer,
FTLD comprises three distinct subtypes: frontotemporal
Demenz (FTD), progressive nonfluent aphasia (PNFA),
and semantic dementia (SD) (Neary et al., 1998). FTD is
associated with unilateral or bilateral prefrontal degen-
eration with personality change, social comportment def-
icits, and impaired self-regulation. SD involves unilateral
or bilateral degeneration of the anterior and inferior
temporal cortex accompanied by a central semantic def-
icit and behavioral changes. PNFA involves left inferior

D 2008 Massachusetts Institute of Technology

Zeitschrift für kognitive Neurowissenschaften 20:10, S. 1839–1853

l

l

/

/

/

/
J

T
T

F
/

ich
T
.

:
/
/

F
R
Ö
M
D
H
Ö
T
w
T
N
P
Ö
:
A
/
D
/
e
M
D
ich
T
F
R
P
Ö
M
R
C
H
.
S
P
ich
l
D
v
ich
e
R
e
R
C
C
T
H
.
M
A
ich
R
e
.
D
C
u
Ö
Ö
M
C
/
N
J
A
Ö
R
C
T
ich
N
C
/
e
A
–
P
R
D
T
ich
2
C
0
l
e
1
0
–
P
1
D
8
F
3
/
9
2
1
0
9
/
3
1
6
0
6
/
1
1
0
8
Ö
3
C
9
N
/
2
1
0
7
0
6
8
4
4
2
8
0
1
2
2
/
6
J
Ö
P
C
D
N
.
B
j
2
0
G
0
u
8
e
.
S
T
2
Ö
0
N
1
2
0
6
7
.
S
P
e
D
P
F
e
M
B
j
B
e
R
G
u
2
0
e
2
S
3
T

/
J

/

F

T

.

.

.

Ö
N

1
8

M
A
j

2
0
2
1

prefrontal degeneration and insidious and progressive
language abnormality, including reduced phrase length,
agrammatism, and effortful and halting speech. Contrast-
ing performance across these subtypes can illuminate
the contribution of damage centered on different brain
regions to autobiographical memory.

In contrast to standard laboratory testing of memory,
assessment of autobiographical memory is affected by
lack of control over encoding and the inherently per-
sonal nature of the memoranda. Autobiographical as-
sessment methods, daher, are more heterogeneous
than laboratory memory tasks, confounding interpreta-
tion of autobiographical memory findings across stud-
ies. In der vorliegenden Studie, we used advanced assessment
methods to assess autobiographical memory in patients
with FTLD. These allowed us to assess several hypoth-
eses concerning the behavioral and neuroanatomical
substrates of autobiographical memory.

The first hypothesis concerned the degree to which
episodic versus semantic autobiographical memory is
affected in FTLD. Episodic autobiographical memory re-
fers to recollection for events specific in time and place,
informa-
with accompanying mental and emotional
tion, including a subjective sense of personal continuity
across time (Tulving, 2002). This latter element is of par-
ticular relevance in FTLD where behavioral disturbances
suggest altered self-awareness (Miller et al., 2001). Se-
mantic autobiographical memory refers to recall of per-
sonal information not specific in time or place, wie zum Beispiel
where one was born, or repeated events, such as yearly
vacations at a particular location. Generally speaking,
studies have shown that patients with FTLD are im-
paired on measures of episodic autobiographical mem-
ory while they are preserved on semantic autobiographical
Erinnerung (Matuszewski et al., 2006; Piolino et al., 2003;
Hodges & Gurd, 1994).

Although episodic and semantic autobiographical
memory occur simultaneously during natural discourse
(Levine, Svoboda, Hay, Winocur, & Moscovitch, 2002), In-
struments for the assessment of autobiographical mem-
ory assume that memories can be wholly classified as
either ‘‘episodic’’ or ‘‘semantic.’’ Thus, these two ele-
ments of autobiographical memory are assessed sepa-
rately using interviews unmatched for content, difficulty,
and psychometric characteristics (z.B., Kopelman, Wilson,
& Baddeley, 1990) or tallied by sorting memories into
episodic or semantic categories (Piolino et al., 2007). Epi-
sodic memory is further quantified on the basis of ordinal
scale ratings that encompass both generic (z.B., repeated
or not temporally specific) and specific autobiographical
events that are subject to contamination by semantic
Verarbeitung.

In this study, we used the Autobiographical Interview
(Levine et al., 2002), which assumes that episodic and
semantic memory reflect distinct information processing
streams that can be simultaneously active during narra-
tive recall. This instrument yields independent, paramet-

ric estimates of episodic and semantic autobiographical
memory derived from within each memory at the time
of scoring rather than with separate interviews at the
time of testing or by wholesale assignment of recalled
narrative events to episodic or semantic categories. Das
measure has proven useful in dissociating episodic from
semantic memory in the elderly (St.-Jacques & Levine, 2007;
Levine et al., 2002),
in patients with medial temporal
lobe damage (Rosenbaum et al., 2008; Addis, Moscovitch,
& McAndrews, 2007; Steinvorth, Levine, & Corkin, 2005),
and in patients with semantic dementia (McKinnon, Black,
Müller, Moscovitch, & Levine, 2006). By quantifying inde-
pendent categories of autobiographical details, the Auto-
biographical Interview permits the assessment of profiles
of autobiographical content, including elevation of non-
episodic details extraneous to the central event as docu-
mented in aging (Levine et al., 2002), semantic dementia
(McKinnon et al., 2006), and frontal dysfunction (Levine,
2004). Based on these studies, we predicted an elevation
of nonepisodic autobiographical content (semantic and
other extraneous details) in FTLD, in addition to a decre-
ment of episodic content. As FTLD is a progressive dis-
Befehl, we assessed this pattern not just cross-sectionally,
but longitudinally, by retesting patients after 1 Zu 2 Jahre.
Decline in episodic, but not semantic, autobiographical
memory over time would suggest that the former is spe-
cifically affected by progressive volume loss in regions af-
fected by FTLD.

The second issue addressed in this study is the role of
retrieval support in autobiographical memory. Retrieval
support bolsters autobiographical memory recall in healthy
Erwachsene (Levine et al., 2002) and in patients with FTLD
(McKinnon et al., 2006; Moss, Kopelman, Cappelletti, Von
Mornay Davies, & Jaldow, 2003; Hodges & Gurd, 1994),
possibly by compensating for executive deficits affecting
strategic search and retrieval of remote information (z.B.,
Craik & McDowd, 1987). Methods of evoking auto-
biographical memories vary widely in the amount of re-
trieval support they provide, confounding comparison of
findings across studies. In this study, we assessed auto-
biographical memory under both low and high levels of
retrieval support to allow direct comparison of the effects
of this manipulation. We also examined the relationship
of neuropsychological test performance to autobiograph-
ical retrieval at these different levels of retrieval support.
To our knowledge, there are no studies of autobiog-
raphical memory in PNFA. Patients with PNFA compose
an interesting comparison group as this condition is less
likely to be associated with disturbances of the self such
as loss of insight and social conduct disorder than are
FTD or SD (Rosen et al., 2006). Given the relationship
between self-related information processing and auto-
biographical memory (Tulving, 2002; Conway & Pleydell-
Pearce, 2000), it was predicted that PNFA patients would
show preservation of episodic autobiographical memory
relative to the other FTLD patients. Andererseits,
the left inferior frontal region affected in PNFA is part

1840

Zeitschrift für kognitive Neurowissenschaften

Volumen 20, Nummer 10

D
Ö
w
N
l
Ö
A
D
e
D

l

l

/

/

/

/
J

T
T

F
/

ich
T
.

:
/
/

F
R
Ö
M
D
H
Ö
T
w
T
N
P
Ö
:
A
/
D
/
e
M
D
ich
T
F
R
P
Ö
M
R
C
H
.
S
P
ich
l
D
v
ich
e
R
e
R
C
C
T
H
.
M
A
ich
R
e
.
D
C
u
Ö
Ö
M
C
/
N
J
A
Ö
R
C
T
ich
N
C
/
e
A
–
P
R
D
T
ich
2
C
0
l
e
1
0
–
P
1
D
8
F
3
/
9
2
1
0
9
/
3
1
6
0
6
/
1
1
0
8
Ö
3
C
9
N
/
2
1
0
7
0
6
8
4
4
2
8
0
1
2
2
/
6
J
Ö
P
C
D
N
.
B
j
2
0
G
0
u
8
e
.
S
T
2
Ö
0
N
1
2
0
6
7
.
S
P
e
D
P
F
e
M
B
j
B
e
R
G
u
2
0
e
2
S
3
T

/
J

T

.

.

F

/

.

Ö
N

1
8

M
A
j

2
0
2
1

of a core autobiographical network in the functional
neuroimaging literature (Svoboda, McKinnon, & Levine,
2006). Given this region’s association with strategic re-
trieval operations (Simons & Spiers, 2003), deficits in
PNFA patients, if present, should be limited to the low
retrieval support condition that places greater demands
on strategic retrieval.

Endlich, we investigated individual differences in auto-
biographical memory performance as a factor of regional
atrophy quantified from patients’ high-resolution struc-
tural magnetic resonance imaging (MRT) scans con-
current to testing. Such data can provide important
adjunctive information to functional neuroimaging stud-
ies by identifying regions necessary for autobiographical
Erinnerung, as opposed to those that are simply engaged
by it. Few studies have assessed quantified cortical dam-
age effects on remote autobiographical memory, mit
results implicating regions across the cortical mantle
(z.B., McKinnon et al., 2007; Eustache et al., 2004;
Kopelman et al., 2003; Eslinger, 1998; Rubin & Greenberg,
1998),
lobe structures
(Rosenbaum et al., 2008; Moscovitch et al., 2005), al-
though this is disputed (Squire & Bayley, 2007).

including medial

zeitlich

In FTLD, the integrity of the medial temporal lobes
has been related to mnemonic processing (So¨derlund,
Black, Müller, Freedman, & Levine, 2008; Simons, Graham,
et al., 2002; Simons, Verfaellie, et al., 2002). Very general
evidence in support of a role for the frontal lobes has
been derived from neuropsychological test data and
diagnostic grouping (Matuszewski et al., 2006; Simons,
Verfaellie, et al., 2002). In a recent study of frontal-variant
FTD patients, florodeoxyglucose (FDG) uptake in the left
orbito-frontal and anterior temporal cortex was correlated
with autobiographical specificity (Piolino et al., 2007).

In der vorliegenden Studie, we used multivariate statistical
analyses of quantitated regional volumes derived from
structural MRI taken concurrent to testing to identify pat-
terns of regional volume loss across the brain associated
with indices of episodic and semantic autobiographical
memory in FTLD. In line with evidence supporting a role
for the medial temporal lobes and associated structures in
episodic autobiographical memory retrieval (Svoboda
et al., 2006; Moscovitch et al., 2005), we predicted that
volume loss in these regions would be strongly and spe-
cifically associated with lifespan retrieval of episodic
autobiographical details. We also predicted distributed
involvement of other regions (particularly frontal and
posterior regions given the strategic and visuospatial ele-
ments of autobiographical memory).

METHODEN

Subjects

FTLD Patients

Patients were recruited from dementia clinics at Baycrest
Centre and Sunnybrook Health Sciences Centre in Tor-

onto and at the University of California at San Francisco
Medical Center. FTLD diagnosis followed the consensus
criteria set out by Neary et al. (1998), which delineates
FTD, PNFA , and SD. Patients with significant aphasia,
neglect, or other focal neurological disturbance or se-
vere cognitive or physical disability that interfered with
testing were excluded. PNFA patients were in the early
stages of disease and, daher, had sufficient residual
speech capacity to participate.

We tested 22 Patienten. Eight met criteria for FTD and
five met criteria for PNFA. Nine patients met criteria for
both FTD and SD (Snowden, Neary, & Mann, 2007) Und
are therefore designated as FTD/SD. Two patients meet-
ing criteria for SD have been reported as part of a sep-
arate study (McKinnon et al., 2006). To date, 10 Patienten
in our sample have come to autopsy. These cases con-
firmed the presence of pathology consistent with FTLD,
including ubiquitin-positive, tau-negative inclusions with
or without degeneration of the motor neurons in some
Fälle, tau-positive Pick bodies inclusions with or with-
out a-synuclein inclusions, or progressive supranuclear
palsy and cortical basal degeneration, which can also
manifest as tauopathy in other cases (McKhann et al.,
2001). There were no significant differences across
groups for age, Ausbildung, duration of illness, or Mini
Mental State Examination (siehe Tabelle 1). Seventeen of
our patients (6 FTD, 7 FTD/SD, Und 4 PNFA) received
high-resolution structural MRI scans as part of the test-
ing protocol. The background characteristics of these
patients were similar to the full sample. Follow-up test-
ing was conducted on eight patients (5 FTD, 1 FTD/SD,
Und 2 PNFA) 1 Zu 2 years after initial assessment. Seven
of these eight patients received repeat MRI scans at the
same time as their follow-up testing.

Comparison Groups

A comparison group for behavioral assessment was
zusammengesetzt aus 16 healthy adults with no history of neu-
rological or psychiatric illness and free from medica-
tion known to affect cognitive functioning. They were
matched to the FTLD patients in terms of age (M =
58, SD = 9.2) and education (M = 16.4, SD = 3.0).
A separate scanning comparison group (n = 10), Auch
matched to the FTLD patients in terms of age (M = 63,
SD = 10.5) and education (M = 17, SD = 3.6), was used
for the purposes of assessing volume reductions in our
FTLD sample and for constructing an MRI template (sehen
below). These subjects were not tested behaviorally and
they were not included with the patients in the brain–
behavior analyses. Exclusion criteria included prior neu-
rological or systemic disease that could affect cogni-
tion, prior psychiatric hospitalization or treatment with
psychiatric medication for greater than 6 weeks, prior
significant alcohol/drug abuse, and significant develop-
mental disabilities.

McKinnon et al.

1841

D
Ö
w
N
l
Ö
A
D
e
D

l

l

/

/

/

/
J

T
T

F
/

ich
T
.

:
/
/

F
R
Ö
M
D
H
Ö
T
w
T
N
P
Ö
:
A
/
D
/
e
M
D
ich
T
F
R
P
Ö
M
R
C
H
.
S
P
ich
l
D
v
ich
e
R
e
R
C
C
T
H
.
M
A
ich
R
e
.
D
C
u
Ö
Ö
M
C
/
N
J
A
Ö
R
C
T
ich
N
C
/
e
A
–
P
R
D
T
ich
2
C
0
l
e
1
0
–
P
1
D
8
F
3
/
9
2
1
0
9
/
3
1
6
0
6
/
1
1
0
8
Ö
3
C
9
N
/
2
1
0
7
0
6
8
4
4
2
8
0
1
2
2
/
6
J
Ö
P
C
D
N
.
B
j
2
0
G
0
u
8
e
.
S
T
2
Ö
0
N
1
2
0
6
7
.
S
P
e
D
P
F
e
M
B
j
B
e
R
G
u
2
0
e
2
S
3
T

/
J

F

T

.

.

.

/

Ö
N

1
8

M
A
j

2
0
2
1

Tisch 1. Characteristics of FTLD Patients and Comparison Subjects

N

Ageb

Comparison

FTD

16 (5 M)

58 (9.2)

8 (5 M)

59 (6.0)

FTD/SD

9 (2 M)

59 (9.4)

PNFA

5 (3 M)

66 (10.4)

Education (Jahre)

16.4 (3.0)

15.5 (4.0)

16.5 (3.0)

17.3 (3.0)

Duration of illness (Jahre)

MMSE

N/A

N/A

3.0 (2–10)

4.0 (2–6)

2.5 (1–5)

25.9 (3.3)

28.0 (1.9)

27.8 (1.5)

Follow-up FTLDa

8 (5 M)

61 (6.2)

15.1 (3.8)

3.0 (2–7)

24.4 (3.4)

MMSE = Mini Mental State Examination.
aThese patients comprise five FTD, one FTD/SD, and two PNFA patients that were retested 1 Zu 2 years after their initial assessment.
bStatistics are means and standard deviations, except for duration of illness, where median and range are reported.

Verfahren

The Autobiographical Interview

Event selection and instructions. The Autobiographi-
cal Interview was administered as described by Levine
et al. (2002), with slight modifications. Subjects were
asked to provide a detailed description of a significant
personal event from each of five life periods: early child-
hood (to age 11 Jahre), teenage years (ages 11–17 years),
early adulthood (ages 18–35 years), middle age (35–
55 Jahre), and the past year. Subjects were instructed to
recall an event that occurred at a specific time and place.
In cases where subjects were unable to generate a specific
life events (z.B.,
event independently, a list of typical
losing something important) was presented to assist in
event retrieval.

In order to examine
Conditions of retrieval support.
facilitative effects of retrieval support on memory, Wir
manipulated the level of structure available to subjects
across three conditions: recall, general probe, and spe-
cific probe (Levine et al., 2002). At recall, subjects spoke
about the event extemporaneously without any inter-
ruption from the examiner, continuing until it was evi-
dent that they had reached a natural ending point. Nach
an event was recalled, general probes were used to clar-
ify instructions and to encourage greater recall of details.
If general probing did not elicit a specific event, Die
subject was given the option of selecting a different
event that was more likely to result in successful recall.
General probes were limited to nonspecific statements
or repetitions of the instructions. At the specific probe
Phase, a structured interview was administered that was
designed to elicit additional contextual details. In order
to prevent the specific probe process from contaminat-
ing recall of subsequent memories, specific probing was
administered after all five events were recounted under
the recall and general probe conditions. Subjects’ de-
scriptions of the selected events were audio-recorded
for later transcription and analysis.

Following probing, subjects were asked to rate the fol-
lowing on a 6-point scale: importance (both at the time
of the event and at the time of testing), visualization,

experienced emotion at the time of the event, and fre-
quency of reactivation (thinking or talking about the
Ereignis). These ratings were unavailable for five FTD pa-
tients. There were no significant group differences in
Bewertungen; these data will not be discussed further.

Text segmentation and categorization. Following
transcription, each memory was segmented into infor-
mational bits or details. Each detail was then classified
according to the procedure outlined in Levine et al.
(2002). Briefly, details were defined as ‘‘internal’’ or epi-
sodic and assigned to one of five categories (Ereignis, Ort,
Zeit, perceptual, and emotion/thought) if they were
related directly to the main event described, were spe-
cific to time and place, and conveyed a sense of episodic
re-experiencing. Ansonsten, details were considered ‘‘ex-
ternal,’’ and consisted of semantic facts (factual infor-
mation or extended events that did not require
recollection of a specific time and place), autobiographical
events tangential or unrelated to the main event, repe-
titions, or other metacognitive statements (‘‘I can’t re-
member.’’) or editorializing (‘‘It was the best of times.’’).
Details were tallied for each category and summed to
form internal and external composites, which were the
main variables of interest in the present study. Scoring
was done separately for each condition (recall, allgemein
probe, specific probe), but scores were analyzed cumu-
latively across levels of recall with general probe and
specific probe details added to details generated from
the prior condition.

To avoid bias in scoring, subjects’ memories were placed
in a common pool and scored at random by seven expe-
rienced scorers who had achieved high interrater reliability
(see Levine et al., 2002) and who were blind to group.

Neuropsychological test results are presented in Table 2.
All neuropsychological tests showed a significant effect
of group (Fs ranged from 4.1 Zu 11.3, ps ranged from
<.001 to <.05). The pattern of group differences was consistent with expectation given the literature on neu- ropsychological differences across subgroups of patients with FTLD. Patients with FTD were impaired on tests of executive functioning and inventories of behavioral change due to frontal damage while being preserved on 1842 Journal of Cognitive Neuroscience Volume 20, Number 10 D o w n l o a d e d l l / / / / j f / t t i t . : / / f r o m D h o t w t n p o : a / d / e m d i t f r p o m r c h . s p i l d v i e r e r c c t h . m a i r e . d c u o o m c / n j a o r c t i n c / e a - p r d t i 2 c 0 l e 1 0 - p 1 d 8 f 3 / 9 2 1 0 9 / 3 1 6 0 6 / 1 1 0 8 o 3 c 9 n / 2 1 0 7 0 6 8 4 4 2 8 0 1 2 2 / 6 j o p c d n . b y 2 0 g 0 u 8 e . s t 2 o 0 n 1 2 0 6 7 . S p e d p f e m b y b e r g u 2 0 e 2 s 3 t / j f . . t / . o n 1 8 M a y 2 0 2 1 Table 2. Neuropsychological Test Performancea FTD (n = 8) FTD/SD (n = 9) PNFA (n = 5) Comparison (n = 16) Trail Making Test, Part Ab Trail Making Test, Part B Phonemic word list generation (FAS) WCSTc perseverations WCST categories Dysexecutive questionnaired Frontal Behavior Inventorye Pyramids and Palm Treesf Boston Naming Testg 11.4 (6.7) 5.4 (3.6)** 9.3 (2.4) 29 (11)* 59 (48)* 5 (2.5)* 24 (11)* 49 (34) 2.8 (4.0)* 2.2 (2.5)* 23 (20)* 43 (10.8) 28 (22)* 32 (13) 47.3 (57.2) 39 (12)**** 50 (9) 31 (14.5)**** 9.6 (3.7) 9.8 (4.9) 25 (13)* 26 (14) 6.2 (2.5) 12.3 (2.2) 10.4 (3.5) 46 (16) 21 (10) 6.9 (3.1) (cid:2)4.4 (10.7)*** (cid:2)3.2 (7.8) 11 (1.8)*** 51 (1) 54 (7) N/A N/A N/A aMeans and standard deviations are reported. bFour patients received the Trail Making Test from the Delis–Kaplan Executive Function System (D-KEFS; Delis, Kaplan, & Kramer, 2001). The remaining patients and comparison subjects received the standard Trail Making Test (Army Individual Test Battery, 1944). All scores were converted to standard scores using the MOANS normative data (Ivnik, 1996) or the D-KEFS manual. Trails A: Control, n = 15. Trails B, FTD, n = 7. cWisconsin Card Sorting Test (Milner, 1963). dBurgess et al., 1996. Other score minus self score (higher scores indicate greater executive problems with less insight). eFrontal Behavior Inventory (Kertesz, Davidson, & Fox, 1997). fPyramids and Palm Trees—picture version. FTD/SD: n = 4. gBoston Naming Test—when necessary, scores were prorated from 15- or 30-item version. FTD, n = 5. FTD/SD, n = 6. PNFA, n = 4. *Significantly different from comparison group. **Significantly different from comparison and FTD groups. ***Significantly different from FTD and FTD/SD groups. ****Significantly different from FTD and PNFA. tasks of semantic retrieval and naming. FTD/SD patients were also impaired on executive tasks and behavioral in- ventories, with additional evidence of semantic and lexi- cal retrieval deficits (although these data are limited by low n) and speeded information processing deficits. PNFA patients show low verbal fluency and preservation on executive tasks and behavioral inventories. Statistical Analyses FTLD patients’ data for certain Autobiographical Inter- view indices were markedly positively skewed. This skew- ness could not be corrected through transformation. We therefore applied a Winsorization procedure to the data by which outliers (i.e., scores exceeding 1.5(cid:1) the intra- quartile range above the third quartile or below the first quartile) were rescaled to be ±2.5 SD from the mean (calculated excluding outliers), allowing the mainte- nance of extreme observations without unduly influenc- ing statistical estimates or sample sizes. We next examined detail composite production across each of the five time periods tested using a 4 (cid:1) 2 (cid:1) 5 mixed-design analysis of variance that treated group (FTD, FTD/SD, PNFA, and comparison) as a between- subjects factor and detail composite (internal, external) and lifetime period (early childhood, teenage years, early adulthood, middle age and past year) as within-subjects factors. There were no interactions involving the internal and external composites when these were incorporated into the lifetime period analyses. We found main effects of lifetime period reflecting a recency effect (i.e., greater recall from the past year; Rubin & Schulkind, 1997). As this effect did not interact with group, it will not be discussed further (for a similar finding, see Piolino et al., 2003). A second analysis examined performance on the category-specific measures of autobiographical retrieval using a 4 (cid:1) 9 mixed-design analysis of variance that treated group (FTD, FTD/SD, PNFA, and comparison) as a between-subjects factor and category (internal event, time, place, perceptual and emotion/thought details; external event, other, and semantic details and repeti- tions) as a within-subjects variable. To examine inter- val change in Autobiographical Interview performance among patients retested after 1 to 2 years, we used a 2 (cid:1) 2 repeated measures design treating detail type (in- ternal, external) and test session (Session 1, Session 2) as within-subjects factors. Tukey’s Honestly Significant Difference post hoc test was used for follow-up pairwise comparisons of between- subjects variables. Where required, the Greenhouse–Geisser correction was applied to effects involving repeated mea- sures. All analyses were performed with alpha set at .05. McKinnon et al. 1843 D o w n l o a d e d l l / / / / j t t f / i t . : / / f r o m D h o t w t n p o : a / d / e m d i t f r p o m r c h . s p i l d v i e r e r c c t h . m a i r e . d c u o o m c / n j a o r c t i n c / e a - p r d t i 2 c 0 l e 1 0 - p 1 d 8 f 3 / 9 2 1 0 9 / 3 1 6 0 6 / 1 1 0 8 o 3 c 9 n / 2 1 0 7 0 6 8 4 4 2 8 0 1 2 2 / 6 j o p c d n . b y 2 0 g 0 u 8 e . s t 2 o 0 n 1 2 0 6 7 . S p e d p f e m b y b e r g u 2 0 e 2 s 3 t / j t / . . . f o n 1 8 M a y 2 0 2 1 Correlations between neuropsychological assessment test scores and performance on the internal and exter- nal detail composites were assessed nonparametrically (Spearman’s (cid:1) [rho]) due to nonnormality of the data. MRI Scan Acquisition Patients and comparison subjects were scanned on 1.5-T scanners (Toronto: Signa, General Electric Medical Sys- tems, Waukesha, WI; San Francisco: Magnetom VISION system, Siemens, Iselin, NJ ) with similar in-plane reso- lution. The Toronto protocol involved axial acquisitions using spoiled gradient-echo T1-weighted 3-D volume im- aging (TR/TE/flip angle = 35 msec/5 msec/358, 1.0 NEX, acquisition matrix = 256 * 256; 124 slices, slice thickness = 1.3 mm; FOV = 22 cm), as well as spin echo, proton density, and T2-weighted images (TR/TE = 3000 msec/ 30 msec, 80 msec, 0.5 NEX, acquisition matrix 256 (cid:1) 192, slice thickness = 3 mm; FOV = 20 cm). The San Francisco protocol also applied a double spin-echo sequence (TR/ TE1/TE2 = 5000/20/85 msec, 51 contiguous 3 mm axial slices covering the entire brain and angulated (cid:2)108 from the AC–PC line). Volumetric T1-weighted gradient-echo MRI were achieved using the MP-RAGE sequence (TR/TE/ TI = 10/4/300 msec, 158 flip angle, 1.5 mm slab thickness) in coronal orientation perpendicular to the double spin- echo sequence. Image Processing Brain MRI data were analyzed via an updated version of our previously reported image processing pipeline (Dade et al., 2004; Kovacevic et al., 2002). The main modification to this protocol involves template match- ing, allowing for comparison of individual images to a standard image and facilitating automation of previously semi-automated steps. The first step in the pipeline was to create an unbiased nonlinear average of T1-weighted images from the age- and education-matched scanning comparison group using a modification of an algorithm previously developed for mouse brain MRI (Kovacevic et al., 2005). Each subject’s T1-weighted image was then registered to the template brain (Woods, Grafton, Holmes, Cherry, & Mazziotta, 1998; Woods, Grafton, Watson, Sicotte, & Mazziotta, 1998) preserving the orig- inal size of the brain while standardizing the position and orientation. Images were resampled into template space using windowed sinc interpolation. Template matching was accomplished via nonlinear registration of T1-weighted images to the template image (Collins & Evans, 1997). Removal of nonbrain tissue from the image incorporated thresholding information derived from the PD- and T2-weighted images, facilitating the distinction between the dura mater and the gray matter (Kovacevic et al., 2002). This is contrasted to methods of brain extraction on the basis of the T1-weighted image that emphasize the cortical surface, inconsistently preserv- ing subdural cerebrospinal fluid (CSF). The voxels on the T1-image were then classified as representing gray matter, white matter, or CSF using an automated tissue classifica- tion method that corrects for radio-frequency inhomoge- neity inherent to MR scanning (Kovacevic et al., 2002). A modified Semi-Automated Brain Region Extraction (SABRE) (Dade et al., 2004) method was then used to create ROIs on the template brain. This method involves manual identification of 15 landmarks and tracing of the cingulate gyrus on the template brain. Based on iden- tification of the edges of the brain and the anterior and posterior commissures, a Talairach-like (Talairach & Tournoux, 1988) grid is automatically created. The algorithm uses this grid along with the landmark co- ordinates to divide the brain into 38 regions (19 per hemisphere; see Figure 1). Nonlinear deformation field matching of the template to individual images was used to customize these regions to fit each subject’s brain anatomy (as opposed to transforming images to fit the template, which can distort interindividual topographi- cal variability). Regional gray matter, white matter, and CSF volumes were adjusted for total intracranial capacity using a regression-based method (Arndt, Cohen, Alliger, Swayze, & Andreasen, 1991). As our segmentation pro- tocol is flexible across different T1-weighted contrasts, no adjustment was necessary to accommodate images acquired from different scanners. Our tissue compartment segmentation and SABRE software are particularly well suited to analysis of brains with atrophy, as they do not require spatial transforma- tion that can distort interindividual topographical vari- ability. They have thus been successfully applied across a variety of patient samples (e.g., Levine et al., 2008; Bocti, Rockel, Roy, Gao, & Black, 2006; Gilboa et al., 2005; Feinstein et al., 2004). Image Analysis Partial least squares (PLS) is a flexible multivariate tech- nique that has been extensively applied to brain imag- ing data (McIntosh, Chau, & Protzner, 2004; McIntosh, Bookstein, Haxby, & Grady, 1996). In general terms, PLS is a multivariate analysis technique for relating two sets of variables to each other. In the present application, it was used to identify patients’ patterns of volume loss re- lated to measures derived from the Autobiographical Interview. It is unbiased in that there are no a priori as- sumptions about structure–function correlations. Because PLS considers the brain as a whole, it is well suited to the detection of distributed patterns of volume loss that co- vary with test performance. In the first step of the PLS analyses, correlations were computed between the brain imaging data (i.e., regional gray matter, white matter, and CSF volumes) for the full sample of patients (without respect to diagnosis) and Autobiographical Interview category scores (i.e., five in- ternal detail categories and four external detail catego- 1844 Journal of Cognitive Neuroscience Volume 20, Number 10 D o w n l o a d e d l l / / / / j f / t t i t . : / / f r o m D h o t w t n p o : a / d / e m d i t f r p o m r c h . s p i l d v i e r e r c c t h . m a i r e . d c u o o m c / n j a o r c t i n c / e a - p r d t i 2 c 0 l e 1 0 - p 1 d 8 f 3 / 9 2 1 0 9 / 3 1 6 0 6 / 1 1 0 8 o 3 c 9 n / 2 1 0 7 0 6 8 4 4 2 8 0 1 2 2 / 6 j o p c d n . b y 2 0 g 0 u 8 e . s t 2 o 0 n 1 2 0 6 7 . S p e d p f e m b y b e r g u 2 0 e 2 s 3 t / j . t . / . f o n 1 8 M a y 2 0 2 1 D o w n l o a d e d l l / / / / j t t f / i t . : / / f r o m D h o t w t n p o : a / d / e m d i t f r p o m r c h . s p i l d v i e r e r c c t h . m a i r e . d c u o o m c / n j a o r c t i n c / e a - p r d t i 2 c 0 l e 1 0 - p 1 d 8 f 3 / 9 2 1 0 9 / 3 1 6 0 6 / 1 1 0 8 o 3 c 9 n / 2 1 0 7 0 6 8 4 4 2 8 0 1 2 2 / 6 j o p c d n . b y 2 0 g 0 u 8 e . s t 2 o 0 n 1 2 0 6 7 . S p e d p f e m b y b e r g u 2 0 e 2 s 3 t / j . / t . . f o n 1 8 M a y 2 0 2 1 Figure 1. SABRE regional cortical divisions in axial and sagittal views. LSF = lateral superior frontal; MSF = medial superior frontal; LMF = lateral middle frontal; MMF = medial middle frontal; LVF = lateral ventral frontal; MVF = medial ventral frontal; GCG = genual cingulate gyrus; ACG = anterior cingulate; MCG = middle cingulate gyrus; PCG = posterior cingulate gyrus/retrosplenial; AT = anterior temporal; MT = medial temporal; PT = posterior temporal; O = occipital; ABGT = anterior basal ganglia/thalamus; PBGT = posterior basal ganglia/ thalamus; EC = external capsule/corona radiata; IP = inferior parietal; SP = superior parietal. ries) for both recall and specific probe, collapsed across time period. The goal of this analysis was to assess brain–behavior correlations in FTLD by treating quanti- fied regional brain atrophy as an independent variable, rather than diagnosis, which is an imprecise proxy for regional brain changes. Singular value decomposition was then applied to the correlation matrix to identify latent variables [LVs] that indicated optimal relations between patterns of regional brain volume loss and test performance. Saliences (sim- ilar to factor loadings in factor analyses) reflected the contribution of individual brain volumes and behavioral measures to the LV (McIntosh et al., 1996, 2004). Mul- tiplication of each region’s salience with its volume and summing over all volumes gives a ‘‘brain score’’ for each patient on a given LV that indicates the degree to which the pattern of volumetric changes identified by the LV is expressed in each patient. Similarly, each patient’s con- tribution to the behavioral aspect of the LV (‘‘behavior score’’) is derived by multiplying the salience for each test by the patient’s test score. The relationship be- tween the two aspects of the LV (patterns of regional brain volume loss and test performance) can be exam- ined by plotting the brain scores against the behavior scores on a patient-by-patient basis, allowing us to ex- amine the contribution of diagnosis to the pattern of brain–behavior relationships (see Figure 5C). The statistical significance of each LV was assessed by 1500 permutation tests (Edgington, 1980), in which the observations are randomly reordered without replace- ment to calculate the probability of each LV having oc- curred by chance. The stability of each brain region’s salience contribution to the LV was determined through bootstrap resampling (subjects were resampled 500 times; McIntosh et al., 1996; Wasserman & Bockenholt, 1989). Brain regions in the singular images were considered reliable if they had a ratio of salience to standard error (hereafter referred to as the boostrap ratio), interpreted similar to a Z-score (McIntosh et al., 2004; Efron & Tibshirani, 1986) greater than 3, corresponding to 99.9% confidence limits. The bootstrap procedure yields 99% confidence intervals around the correlations between test scores and the pattern of regional brain volume changes. Because image-wide statistical assessment is done in a sin- gle analytic step, no correction for multiple comparisons across brain regions is required. For the purposes of characterizing the degree of brain atrophy in our patient sample, the 19 SABRE regions per hemisphere were reduced to eight (ventral frontal, dorsal frontal, anterior, medial, and posterior temporal, inferior parietal, superior parietal, and occipital). Reduction in parenchymal volumes (gray + white matter) were as- sessed via planned comparisons between the FTD, FTD/ SD, and PNFA subgroups and the 10 matched scanning comparison subjects. Interval change between the first and second scans for the seven rescanned FTLD patients was also assessed via planned comparisons. RESULTS Patterns of Brain Atrophy in FTLD Patients frontal regions bilaterally, FTLD patients showed a characteristic pattern of signif- icantly reduced parenchymal volumes maximal over the frontal and temporal lobes. For illustration, the left dor- solateral frontal and posterior temporal volumes are displayed in Figure 2. The FTD group had volume loss over the dorsolateral left ( p < .009) greater than right ( p < .05), as well as sig- nificant volume loss in the right posterior temporal re- gion ( p < .005). The FTD/SD group had marked volume loss over anterior, posterior, and medial temporal lobe sectors ( ps ranging from .0001 to .02) as well as all fron- tal sectors ( ps ranging from .001 to .007). Volume loss in PNFA patients was more restricted, maximal in the left dorsolateral region ( p < .001), but also in the right pos- terior temporal and right inferior parietal regions ( ps < .05). It should be noted that only four PNFA patients McKinnon et al. 1845 of autobiographical memory. A significant Detail cate- gory (cid:1) Group interaction [F(24, 272) = 7.03, p < .001] was due to lower production of internal details in the FTD and the FTD/SD groups than in comparison subjects for all internal detail categories ( ps < .001–.05) except for time, which was uniquely impaired in the FTD/SD group ( p < .05), and for place, which was uniquely impaired in the FTD group ( p < .05). The FTD/SD group produced more semantic details than did the comparison group ( p < .05). Significant differences between the PNFA pa- tients and comparison subjects were limited to the internal perceptual and thought/emotion categories ( ps < .05). Specific Probe Following specific probing, the significant Detail type (cid:1) Group interaction remained [F(3, 128) = 8.38, p < .001], with greater production of internal details by the com- parison group as compared to the FTD and FTD/SD groups ( p < .01; see Figure 3). Whereas the PNFA group produced significantly fewer internal details in the recall condition than did comparison subjects, after specific probing, the PNFA group’s production of internal details differed neither from the comparison subjects nor from the FTD group. There were again no group differences for external details. The main effect of group was not significant, indicating that all groups generated proto- cols of similar length. The Detail category (cid:1) Group interaction [F(24, 272) = 4.49, p < .001] remained significant at specific probe. Whereas the FTD/SD group produced fewer internal de- tails than did the comparison group for all internal detail categories ( ps < .001–.01) except for time, where this group was unimpaired, the FTD group was impaired rel- ative to the comparison group for the perceptual and thought/emotion internal categories only ( ps < .01). D o w n l o a d e d l l / / / / j f / t t i t . : / / f r o m D h o t w t n p o : a / d / e m d i t f r p o m r c h . s p i l d v i e r e r c c t h . m a i r e . d c u o o m c / n j a o r c t i n c / e a - p r d t i 2 c 0 l e 1 0 - p 1 d 8 f 3 / 9 2 1 0 9 / 3 1 6 0 6 / 1 1 0 8 o 3 c 9 n / 2 1 0 7 0 6 8 4 4 2 8 0 1 2 2 / 6 j o p c d n . b y 2 0 g 0 u 8 e . s t 2 o 0 n 1 2 0 6 7 . S p e d p f e m b y b e r g u 2 0 e 2 s 3 t / j . . f . t / o n 1 8 M a y 2 0 2 1 Figure 3. Mean number of internal (episodic) and external (nonepisodic) details generated by comparison subjects, FTD, FTD/SD, PNFA, on the Autobiographical Interview (specific probe phase). FTD (n = 8) and FTD/SD (n = 9) patients generated fewer internal details and, in the case of FTD/SD, more external (semantic) details, whereas PNFA patients (n = 5) were not significantly different from comparison subjects. *Significantly different from comparison subjects, p < .01. Figure 2. Regional parenchymal volumes in FTLD patients and comparison subjects in left dorsolateral prefrontal and in left posterior temporal regions. Comp. = comparison subjects. Vertical lines indicate 95% confidence intervals. Group means are indicated by bold horizontal lines. Relative to comparison subjects, patients in all groups showed significant decline in left dorsolateral prefrontal volume, whereas only the FTD/SD subgroup had significant volume loss in the left posterior temporal region. were scanned, one of whom was left-handed with pri- marily right-sided degeneration, suggesting crossed lat- eralization of language. Interval change for the seven rescanned FTLD patients was statistically significant over all regions of interest ( ps ranged from .0005 to .03). Autobiographical Memory Recall A significant interaction between detail type and group [F(3, 132) = 14.18, p < .01], during the free recall phase of the Autobiographical Interview, was due to reduced production of internal details in the FTD, FTD/SD and PNFA groups relative to comparison subjects ( ps < .001, .001, and .05 for the FTD, FTD/SD, and PNFA groups, respectively); there were no group differences for exter- nal details. The main effect of group was not significant ( p > .05), indicating that the patients’ reduction in in-
ternal details was not attributable to an overall reduction
in protocol length among the patient groups.

Analysis of individual detail categories provided more
information concerning group differences in the elements

1846

Zeitschrift für kognitive Neurowissenschaften

Volumen 20, Nummer 10

The FTD/SD group continued to produce more seman-
tic details than did the comparison group ( P < .05). Significant differences between the PNFA patients and comparison subjects were limited to the thought/ emotion category ( p < .01). One-year Follow-up Testing Among the eight patients retested 1 year later, there was a significant Test session (cid:1) Detail type interaction at specific probe [F(1, 7) = 8.77, p < .05] that was due to a reduction of internal ( p < .005) details across the test sessions (see Figure 4). External details increased across test sessions, although this difference fell short of statis- tical significance due to high variability. There were no significant main effects involving test session or detail type at recall or specific probe. The Test session (cid:1) Detail type interaction at recall was not significant. Relation of Autobiographical Memory to Neuropsychological Test Scores Both FTLD patients and comparison subjects showed a positive relation between performance on the phonemic word list generation and scores on the internal detail composite at recall [(cid:1) (21, 16) = 0.54 and 0.64, for pa- tients and comparison subjects, respectively, ps < .02 and .01]. There was a correlation between Trail Making, Part A and internal details at recall in FTLD patients [(cid:1) (21) = 0.45, p < .05]. There were no significant cor- relations between neuropsychological test performance and internal details at specific probe. External details at specific probe were negatively correlated with Trail Mak- ing, Part B in patients [(cid:1) (20) = (cid:2)0.47, p < .05]; there were no other significant correlations between neuro- psychological tests and the external detail composites. Figure 4. Effects of FTLD (n = 8) progression on internal (episodic) and external (nonepisodic) autobiographical details from the Autobiographical Interview (specific probe phase). *Significantly different from Session 1, p < .005. Error bars represent standard error of the mean. Relation to Regional Brain Atrophy The PLS analysis allowed us to examine the pattern of regional volume loss associated with patients’ impaired performance on the Autobiographical Interview. One LV was identified ( p = .05) by permutation test. This LV accounted for 41% of the covariance between test scores and regional volumes. The pattern of Autobiographical Interview scores associated with this LV indicated a sepa- ration of episodic from nonepisodic details concentrated at specific probe (see Figure 5A). Specifically, increased event, time, place, and perceptual details at specific probe and event details at recall were related to increased parenchymal volumes across the identified brain regions, whereas increased semantic details and repetitions at specific probe were related to decreased volumes. Figure 5B depicts significant negative bootstrap ratios for CSF (i.e., indicating increased parenchymal volume positively associated with internal details and negatively associated with external details) with foci in the left posterior and anterior temporal lobes and the medial temporal lobes bilaterally. The left inferior parietal lobe, the left posterior cingulate/retrosplenial region, and the bilateral posterior basal ganglia regions also emerged as significant in this analysis. There were additional signif- icant positive bootstrap ratios for gray and white matter (not displayed in the figure) that mirrored those de- picted in Figure 5: left posterior temporal and left oc- cipital gray and white matter, bilateral posterior basal ganglia white matter, and left anterior basal ganglia white matter. Significant negative bootstrap ratios were observed for the left anterior cingulate gyrus and bilat- eral internal capsule/corona radiata gray matter regions. In Figure 5C, the brain scores (reflecting each pa- tient’s contribution to the pattern of volume loss iden- tified by the LV) and the behavior scores (reflecting each patient’s contribution to the pattern of test scores iden- tified by the LV) are plotted against each other. Al- though the sample is small and there is overlap across it can be seen that the lower the diagnostic groups, ends of the behavioral and brain score distributions are mainly occupied by patients in the FTD/SD group, whereas FTD and PNFA patients occupy the upper end. DISCUSSION Lifespan autobiographical episodic recall was impaired in patients with FTLD relative to comparison subjects, whereas semantic autobiographical memory was spared or elevated in patients with FTLD. Narrative episodic specificity was associated with the integrity of a left- lateralized posterior network centered on the temporal lobe. The specificity of the episodic relative to semantic impairment was further supported by longitudinal data. When the medial temporal lobes are spared, as in PNFA, retrieval support is effective in increasing autobiograph- ical recall to normal levels. Contrary to expectation, we McKinnon et al. 1847 D o w n l o a d e d l l / / / / j t t f / i t . : / / f r o m D h o t w t n p o : a / d / e m d i t f r p o m r c h . s p i l d v i e r e r c c t h . m a i r e . d c u o o m c / n j a o r c t i n c / e a - p r d t i 2 c 0 l e 1 0 - p 1 d 8 f 3 / 9 2 1 0 9 / 3 1 6 0 6 / 1 1 0 8 o 3 c 9 n / 2 1 0 7 0 6 8 4 4 2 8 0 1 2 2 / 6 j o p c d n . b y 2 0 g 0 u 8 e . s t 2 o 0 n 1 2 0 6 7 . S p e d p f e m b y b e r g u 2 0 e 2 s 3 t / j / . . . t f o n 1 8 M a y 2 0 2 1 Figure 5. LV from PLS analysis indicating the association of performance on the Autobiographical Interview with patterns of regional brain volume changes. (A) The pattern of Autobiographical Interview measures associated with the LV, expressed as correlations between test scores and the pattern of brain volume changes depicted in B. Error bars represent 99% confidence intervals. Measures with error bars crossing the horizontal axis did not significantly contribute to the LV. This panel shows that internal event details at recall and internal event, place, time, and perceptual details at specific probe were positively correlated with the pattern of brain volume changes depicted in B, whereas external semantic and repetition details at the specific probe phase were negatively correlated with the pattern of brain volume changes. Ed = event details (note that event details can be either internal or external, depending on whether they concern the main event); Perc = perceptual details; Tho = thought details; Sem = semantic details; Rep = repetitions; Oth = other details (e.g., metacognitive or editorializing). (B) The pattern of regional CSF volume (expressed as color-coded bootstrap ratios) associated with the pattern of test scores depicted in the top panel. The color bar indicates the coding scheme according to the level of the bootstrap ratio, interpreted similar to a Z-score. Lower values (darker blue) correspond to brain volume decreases (i.e., less CSF, greater parenchyma) associated with higher internal details and lower external details. Images were thresholded at a bootstrap ratio of 3.0, corresponding approximately to p < .001. Axial images are displayed in radiological convention (right hemisphere displayed on left side of image). The right and left cingulate volumes are displayed on the right and left side of the images, respectively. Not pictured: Bootstrap ratios for gray and white matter (see text). (C) A scatterplot of individual patients’ brain scores (indicating the degree to which the pattern of atrophy identified by the LV is expressed in each patient) and behavior scores (indicating the degree to which the pattern of test scores identified by the LV is expressed in each patient), with each patient color-coded according to diagnosis, allowing for the appreciation of the how FTLD diagnosis contributes to the brain–behavior relationships. 1848 Journal of Cognitive Neuroscience Volume 20, Number 10 D o w n l o a d e d l l / / / / j t t f / i t . : / / f r o m D h o t w t n p o : a / d / e m d i t f r p o m r c h . s p i l d v i e r e r c c t h . m a i r e . d c u o o m c / n j a o r c t i n c / e a - p r d t i 2 c 0 l e 1 0 - p 1 d 8 f 3 / 9 2 1 0 9 / 3 1 6 0 6 / 1 1 0 8 o 3 c 9 n / 2 1 0 7 0 6 8 4 4 2 8 0 1 2 2 / 6 j o p c d n . b y 2 0 g 0 u 8 e . s t 2 o 0 n 1 2 0 6 7 . S p e d p f e m b y b e r g u 2 0 e 2 s 3 t / j . t . . / f o n 1 8 M a y 2 0 2 1 found no significant relation of frontal lobe volumes to autobiographical memory retrieval. Overall, these find- ings support a behavioral and neuroanatomical distinc- tion between episodic and semantic autobiographical memory. They further suggest that, contrary to standard diagnostic criteria, memory function can be significantly affected in FTLD. Dissociating Episodic and Semantic Autobiographical Memory across FTLD Diagnoses Prior results indicating impaired episodic autobiograph- ical memory and spared semantic autobiographical memory in FTLD (Piolino et al., 2003; Nestor et al., 2002) are based on the use of ordinal ratings that can be elevated by inclusion of nonepisodic details that are re- lated to the event, but do not require recreation of tem- poral, spatial, and other contextual details. One approach to this issue is to apply a binary transformation to the or- dinal data that divides those receiving the highest ratings (considered ‘‘strictly episodic’’) from those receiving lower ratings (Matuszewski et al., 2006; Piolino et al., 2003). Our approach is to extract estimates of episodic and semantic elements from within each memory, under the assumption that such elements occur simultaneously to varying degrees within all naturalistic autobiographical narratives and that these can be independently affected by normal or pathological intersubject variables (e.g., ag- ing, brain disease; Rosenbaum et al., 2008; Addis et al., 2007; Steinvorth et al., 2005; Levine, 2004; Levine et al., 2002) or by intrasubject manipulations (e.g., retrieval sup- port, emotional state; St.-Jacques & Levine, 2007; Levine et al., 2002). The PNFA group was less impaired than the FTD and FTD/SD groups. Indeed, retrieval support selectively bolstered recall of episodic autobiographical details in patients with PNFA such that they were not statistically differentiated from comparison subjects. Mild episodic autobiographical deficits in patients with PNFA are there- fore likely attributable to strategic retrieval or speech ini- tiation deficits that are ameliorated by retrieval support. Patients with FTD and FTD/SD remained impaired rela- tive to comparison subjects after retrieval support, al- though effects across individual categories of episodic details were more reliable among those with FTD/SD. As described in more detail below, this may be attribut- able to the constraining effects of medial temporal lobe damage on episodic autobiographical recall. These re- sults cannot be attributed to reduced speech output, as there were no effects of overall detail production across groups. Patterns of spared and impaired autobiographical memory across these groups are further clarified by con- sideration of nonepisodic (external) autobiographical details. Although there was no overall effect on external details, detail category analyses revealed that semantic autobiographical details were uniquely elevated in the FTD/SD group, indicating that these patients produce an excess of generic autobiographical information in addi- tion to reduced richness of happenings, spatial referen- ces, perceptual information, and mental states. Patients with PNFA are distinguished from patients with FTD by a relative preservation of insight and social and personal conduct (Neary et al., 1998), an observa- tion confirmed by these patients’ normal scores on in- ventories of behavioral and social cognitive changes. Although deficits in these areas are not core to the diag- nosis of SD in its original formulation, there is evidence that behavioral features in SD and FTD significantly over- lap (Rosen et al., 2006; Bozeat, Gregory, Ralph, & Hodges, 2000). Our findings are consistent with the notion that self-referential processing is fundamental to the capacity for episodic autobiographical memory (Tulving, 2002; Conway & Pleydell-Pearce, 2000), with both processes spared in PNFA relative to FTD and SD. It is acknowledged that interpretation of results may be limited by low power as there were only five patients in the PNFA groups. Yet these patients were differenti- ated from the FTD and FTD/SD groups not just by de- gree, but by their pattern of distribution of details across internal and external categories, which paralleled that of the comparison group. Although there were no signifi- cant group differences for disease duration, PNFA pa- tients may have tested at an earlier stage of disease progression as those with significant aphasia were ex- cluded. We directly assessed the effects of disease pro- gression by retesting patients after 1 to 2 years, where the pattern of reduced episodic but not semantic autobiographical memory was enhanced, reinforcing the specificity of episodic autobiographical memory loss in FTLD. We did not have sufficient data to differentiate disease progression effects according to diagnosis. Given evidence that differences across FTLD diagnostic entities blur with disease progression (Kertesz, McMonagle, Blair, Davidson, & Munoz, 2005), it is likely that the au- tobiographical profile in late PNFA is similar to that of FTD and SD. Relation of Autobiographical Memory Performance to Patterns of Regional Volume Loss For the purposes of analyzing the effects of diagnosis, we sought to define groups with distinct clinical syn- dromes. In our brain–behavior analysis, we directly as- sessed the effects of individual differences in patterns of brain atrophy across clinical diagnoses, rather than inferring lesion location from diagnosis. The dissociation between episodic and semantic autobiographical mem- ory was further reinforced by the brain–behavior analy- ses. Critically, the patterns of regional atrophy that emerged in relation to this dissociation were not spec- ified in a priori planned contrasts (as is typically done in brain–behavior analyses, for example, by contrasting groups with presumed focal patterns of damage), but McKinnon et al. 1849 D o w n l o a d e d l l / / / / j f / t t i t . : / / f r o m D h o t w t n p o : a / d / e m d i t f r p o m r c h . s p i l d v i e r e r c c t h . m a i r e . d c u o o m c / n j a o r c t i n c / e a - p r d t i 2 c 0 l e 1 0 - p 1 d 8 f 3 / 9 2 1 0 9 / 3 1 6 0 6 / 1 1 0 8 o 3 c 9 n / 2 1 0 7 0 6 8 4 4 2 8 0 1 2 2 / 6 j o p c d n . b y 2 0 g 0 u 8 e . s t 2 o 0 n 1 2 0 6 7 . S p e d p f e m b y b e r g u 2 0 e 2 s 3 t / j t / f . . . o n 1 8 M a y 2 0 2 1 rather from unrestricted multivariate analyses of brain– behavior correlations. Regional volumes in a distributed network over the temporal lobes, posterior subcortical regions, and left inferior parietal and occipital regions were positively related to four out of five indices of episodic autobio- graphical memory at specific probe, as well as one (event details) at recall. Volumes in these same regions were negatively related to semantic details and repetitions at specific probe. Thus, in FTLD, specificity and richness of narrative production of lifespan autobiographical memory are related to the integrity of a left-lateralized posterior network centered on the temporal lobes. Our findings are consistent with other patient and functional neuroimaging studies emphasizing the di- verse mnemonic, attentional, and multimodal sensory processes that enable autobiographical re-experiencing (Svoboda et al., 2006; Kopelman et al., 2003). In inter- preting the brain behavior analysis, we emphasize CSF volumes (i.e., the inverse of parenchymal volume). Gray and white matter volume paralleled the CSF effects, al- though not all regions identified in the CSF emerged as significant in the gray and white matter analyses, and some regions (left occipital lobe, left anterior basal gan- glia) emerged in gray and white matter analyses, but not CSF analyses. Although these were consistent with the overall CSF pattern, the left anterior cingulate and in- ternal capsule/corona radiata gray matter showed a re- lationship opposite to the direction of the other regions (i.e., negatively correlated with internal details, posi- tively correlated with external details). As these regions have relatively small volumes (e.g., mean gray matter volume for the left anterior cingulate gyrus: 1777 mm3 vs. 41,166 mm3 for the left posterior temporal lobe), they may be relatively unstable and will not be inter- preted further. Right and left medial temporal lobe volumes were re- lated to autobiographical memory specificity. Our medial temporal lobe region included the hippocampus and surrounding structures, regarded as critical for the bind- ing of features required for recall of contextual details (Eichenbaum, 2000), a distinction that holds within auto- biographical memory (Spiers, Maguire, & Burgess, 2001; Kapur, 1999; Cermak, 1985). The dependence of episodic autobiographical memory on the medial temporal lobes is reinforced by recent studies of patients with focal dam- age in these regions and selective impairment on episodic details as assessed by the Autobiographical Interview (Rosenbaum et al., 2008; Addis et al., 2007; Steinvorth et al., 2005; but see Bright et al., 2006; Kirwan, Bayley, Galvan, & Squire, 2008; and Squire & Bayley, 2007, for different results), as well as functional neuroimaging data (see also Svoboda et al., 2006; Moscovitch et al., 2005). The left posterior cingulate/retrosplenial region was also significantly related to autobiographical memory specificity. This region is connected to the hippocampus and the prefrontal cortex (Morris, Pandya, & Petrides, 1999; Rosene & Van Hoesen, 1977), facilitating integra- tion of mnemonic and higher level cognitive processing. It is also directly connected to multiple thalamic nuclei (Morris, Petrides, & Pandya, 1999) including the anterior thalamus (Vogt, Pandya, & Rosene, 1987) that is part of a midline diencephalic system involved in episodic mem- ory (Aggleton & Brown, 1999). Damage to this region, which is frequently activated in functional neuroimaging studies of autobiographical memory (Svoboda et al., 2006), can cause retrograde amnesia (Valenstein et al., 1987). The left anterior and posterior temporal regions were strongly associated with episodic autobiographical mem- ory retrieval, along with the left inferior parietal region. These regions are also strongly associated with autobio- graphical memory in functional neuroimaging studies (Svoboda et al., 2006) due to retrieval of specific factual information (anterior temporal cortex), semantic pro- cessing (posterior temporal cortex), and visuospatial processing (temporo-parietal junction). Occipital involve- ment (derived from analyses of gray and white matter) is consistent with the importance of visual processes in rec- ollection (Rubin & Greenberg, 1998). The left-lateralization of neocortical activation associated with autobiographical memory and the present findings are inconsistent with evidence of asymmetric right lateral temporal lobe in- volvement in episodic autobiographical memory (Gilboa et al., 2005; Eustache et al., 2004; Kopelman et al., 2003; Eslinger, 1998; for exception, see Piolino et al., 2007). Our findings cannot be explained by asymmetric damage in our patients, as the left and right temporal regions were similarly affected. As an alternative account, it is noted that our findings were concentrated at the specific probe phase as contrasted to prior studies that used the autobiographical memory interview (AMI) (or variants thereof ) that is more comparable to procedures in our recall phase. Conversely, AMI probing of semantic auto- biographical memory, which is highly structured, has been related to left temporal volumes (Gilboa et al., 2005; Eslinger, 1998). Thus, the left-lateralized findings may indicate lack of access to queried specific autobio- graphical details. Our patients with compromised ability to retrieve such details produced an excess of nonspe- cific semantic autobiographical details, reflecting com- pensation, disinhibition, and impaired monitoring of mnemonic output. The semantic autobiographical de- tails elevated in association with left posterior pathology are not equivalent to the central semantic concepts affected by temporal lobe damage in SD. Rather, they reflect autobiographical information not specific in time and place that can be preserved in association with left lateral temporal damage (Westmacott & Moscovitch, 2003; Snowden, Griffiths, & Neary, 1994). The regions related to autobiographical memory per- formance composed only part of the overall pattern of volume loss in our patients, which included the signif- icant frontal lobe changes. Given the established role 1850 Journal of Cognitive Neuroscience Volume 20, Number 10 D o w n l o a d e d l l / / / / j f / t t i t . : / / f r o m D h o t w t n p o : a / d / e m d i t f r p o m r c h . s p i l d v i e r e r c c t h . m a i r e . d c u o o m c / n j a o r c t i n c / e a - p r d t i 2 c 0 l e 1 0 - p 1 d 8 f 3 / 9 2 1 0 9 / 3 1 6 0 6 / 1 1 0 8 o 3 c 9 n / 2 1 0 7 0 6 8 4 4 2 8 0 1 2 2 / 6 j o p c d n . b y 2 0 g 0 u 8 e . s t 2 o 0 n 1 2 0 6 7 . S p e d p f e m b y b e r g u 2 0 e 2 s 3 t / j / . t . f . o n 1 8 M a y 2 0 2 1 of the prefrontal cortex in autobiographical memory (McKinnon et al., 2007; Svoboda et al., 2006; Kopelman et al., 2003), it is surprising these regions did not emerge as related to behavior in our analysis. SABRE-defined frontal volumes also were not related to autobiograph- ical memory performance in patients with Alzheimer’s disease (Gilboa et al., 2005). On the other hand, left orbito-frontal FDG uptake was related to autobiograph- ical memory performance in patients with frontal-variant FTD (Piolino et al., 2007; for a related study in Alzheimer patients, see Eustache et al., 2004), a discrepancy at- tributable to interstudy differences in patient selection, assessment methods, and imaging platform. Although performance on neuropsychological tests of executive functioning has been related to autobiographical mem- ory performance in FTLD (Matuszewski et al., 2006), we found that speeded neuropsychological tests (phonemic word list generation and Trail Making, Part A) were related to episodic autobiographical memory, but not a widely used untimed test of executive functioning, the Wisconsin Card Sorting Test. These correlations were limited to the free recall phase of the Autobiographical Interview. These findings suggest a relation between lexical semantic retrieval and generalized cognitive func- tioning and strategic autobiographical memory retrieval, perhaps due to the cognitive demands involved in cue specification at recall, whereas the structured cueing at the specific probe phase lessens the demands on these processes. At specific probe, the negative relationship between Trail Making, Part B and external details sug- gests that an excess of nonepisodic autobiographical information may be related to higher-level attentional deficits, especially among FTD/SD patients, who per- formed poorly on Trail Making, Part B and had signifi- cantly elevated semantic details. In another study using an overlapping sample of FTLD patients (So¨derlund et al., 2008), performance on labo- ratory tasks of episodic memory, including source mem- ory and remember/know judgments, was associated with the integrity of the left temporal lobe and not the frontal lobes. These findings suggest that although retrieval can be reliably associated with temporal lobe damage in FTLD, the functional localization of advanced states of mnemonic consciousness may be more distributed, without a specific mapping to prefrontal regions, at least in FTLD, although functional changes in frontal regions, possibly reflecting alternations in frontal–posterior net- works, cannot be ruled out as contributing to patients’ altered autobiographical memory function. As noted above, the brain–behavior relationships were most strongly evident at the specific probe phase, where prefrontally mediated executive-retrieval mechanisms might be less in demand (Matuszewski et al., 2006). This may reflect the fact that, in the context of significant temporal lobe damage limiting access to episodic auto- biographical information, prefrontal function does not con- tribute to autobiographical retrieval, even with retrieval support (Rosenbaum, McKinnon, Levine, & Moscovitch, 2004; Kopelman et al., 2003). This also may explain why patients with PNFA, with less medial temporal lobe damage, benefited from retrieval support, whereas those with FTD/SD did not. Conclusion Autobiographical memory has recently received increas- ing attention by cognitive neuroscientists interested in human remote memory processes. Although it is clear that autobiographical memory is multifactorial, with an accordingly distributed functional neuroanatomy, the delineation of specific brain–behavior relationships has been elusive, likely owing to significant heterogeneity in behavioral measures, imaging methods, and neuropsy- chological populations. In this study, FTLD, a form of presenile dementia affecting key regions in the autobiographical memory network, was used as a model for autobiographical memory impairment. With the Autobiographical Inter- view, we localized this impairment to episodic autobio- graphical memory in both cross-sectional and longitudinal analyses. Contrasting FTLD subtypes further specified pat- terns of altered autobiographical memory. Despite their mild speech abnormalities, PNFA patients were the least impaired and were unique among subgroups in their ca- pacity to benefit from retrieval support. FTD/SD patients were the most impaired, failed to benefit from retrieval support, and generated an excess of semantic autobio- graphical details. FTD patients occupied a middle posi- tion, impaired episodic, but not semantic autobiographical memory, and a failure to benefit from retrieval support. The identified pattern of impaired episodic but not se- mantic autobiographical memory was related to volume loss in a distributed network over the temporal lobes, posterior subcortical regions, and left inferior parietal and occipital regions. The capacity of patients with PNFA to benefit from specific cueing may relate to these pa- tients’ relative sparing of the medial temporal regions. The lack of relation between frontal lobe volumes and autobiographical memory performance suggests that this brain–behavior correlation may be more complex than previously recognized. Acknowledgments We thank the patients and their families for their assistance, and Ann Campbell, Marina Mandic, Charlene O’Connor, Colleen O’Toole, Joel Ramirez, Adriana Restagno, Gary Turner, and Eva Svoboda for technical assistance. This research was supported by grants from the Canadian Institutes of Health Research (Grant nos. MT-14744, MOP-37535, and MOP-108540 to B. L., and MT-13129 to S. E. B.), and the NIH-NICHD (Grant no. HD42385-01) to B. L. Reprint requests should be sent to Brian Levine, Rotman Research Institute, Baycrest Centre, 3560 Bathurst Street, Toronto, ON, Canada M6A 2E1, or via e-mail: blevine@rotman-baycrest.on.ca. McKinnon et al. 1851 D o w n l o a d e d l l / / / / j t t f / i t . : / / f r o m D h o t w t n p o : a / d / e m d i t f r p o m r c h . s p i l d v i e r e r c c t h . m a i r e . d c u o o m c / n j a o r c t i n c / e a - p r d t i 2 c 0 l e 1 0 - p 1 d 8 f 3 / 9 2 1 0 9 / 3 1 6 0 6 / 1 1 0 8 o 3 c 9 n / 2 1 0 7 0 6 8 4 4 2 8 0 1 2 2 / 6 j o p c d n . b y 2 0 g 0 u 8 e . s t 2 o 0 n 1 2 0 6 7 . S p e d p f e m b y b e r g u 2 0 e 2 s 3 t / j . f / t . . o n 1 8 M a y 2 0 2 1 REFERENCES Addis, D. R., Moscovitch, M., & McAndrews, M. P. (2007). Consequences of hippocampal damage across the autobiographical memory network in left temporal lobe epilepsy. Brain, 130, 2327–2342. Aggleton, J. P., & Brown, M. W. (1999). Episodic memory, amnesia, and the hippocampal–anterior thalamic axis. Behavioral and Brain Sciences, 22, 425–444; discussion 444–489. Arndt, S., Cohen, G., Alliger, R. J., Swayze, V. W., II, & Andreasen, N. C. (1991). Problems with ratio and proportion measures of imaged cerebral structures. Psychiatry Research, 40, 79–89. Army Individual Test Battery. (1944). Manual of directions and scoring. Washington, D.C.: War Department, Adjutant General’s Office. Bocti, C., Rockel, C., Roy, P., Gao, F., & Black, S. E. (2006). Topographical patterns of lobar atrophy in frontotemporal dementia and Alzheimer’s disease. Dementia and Geriatric Cognitive Disorders, 21, 364–372. Bozeat, S., Gregory, C. A., Ralph, M. A., & Hodges, J. R. (2000). Which neuropsychiatric and behavioural features distinguish frontal and temporal variants of frontotemporal dementia from Alzheimer’s disease? Journal of Neurology, Neurosurgery and Psychiatry, 69, 178–186. Bright, P., Buckman, J., Fradera, A., Yoshimasu, H., Colchester, A. C., & Kopelman, M. D. (2006). Retrograde amnesia in patients with hippocampal, medial temporal, temporal lobe, or frontal pathology. Learning and Memory, 13, 545–557. Burgess, P. W., Alderman, N., Evans, J. J., Wilson, B. A., & Emslie, H. (1996). The Dysexecutive Questionnaire. In B. A. Wilson, N. Alderman, P. W. Burgess, H. Emslie, & J. J. Evans (Eds.), Behavioral assessment of the dysexecutive syndrome. Bury St. Edmunds: Thames Valley Test Company. Cermak, L. S. (1985). The episodic–semantic distinction in amnesia. In L. R. Squire & N. Butters (Eds.), Neuropsychology of memory (pp. 55–62). New York: Guilford Press. Collins, D., & Evans, A. (1997). ANIMAL: Validation and applications of non-linear registration based segmentation. International Journal of Pattern Recognition, 11, 1271–1294. Conway, M. A., & Pleydell-Pearce, C. W. (2000). The construction of autobiographical memories in the self- memory system. Psychological Review, 107, 261–288. Craik, F. I. M., & McDowd, J. M. (1987). Age differences in recall and recognition. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13, 474–479. Dade, L. A., Gao, F. Q., Kovacevic, N., Roy, P., Rockel, C., O’Toole, C. M., et al. (2004). Semiautomatic brain region extraction: A method of parcellating brain regions from structural magnetic resonance images. Neuroimage, 22, 1492–1502. Delis, D. C., Kaplan, E., & Kramer, J. H. (2001). Delis-Kaplan Executive Functioning System (D-KEFS). San Antonio: The Psychological Corporation. Edgington, E. S. (1980). Randomization tests. New York: Marcel Dekker. Efron, B., & Tibshirani, R. (1986). Bootstrap methods for standard errors, confidence intervals and other measures of statistical accuracy. Statistical Sciences, 1, 54–77. Eichenbaum, H. (2000). A cortical–hippocampal system for declarative memory. Nature Reviews Neuroscience, 1, 41–50. Eslinger, P. J. (1998). Autobiographical memory after temporal lobe lesions. Neurocase, 4, 481–495. Eustache, F., Piolino, P., Giffard, B., Viader, F., De La Sayette, V., Baron, J. C., et al. (2004). ‘‘In the course of time’’: A PET study of the cerebral substrates of autobiographical amnesia in Alzheimer’s disease. Brain, 127, 1549–1560. Feinstein, A., Roy, P., Lobaugh, N., Feinstein, K., O’Connor, P., & Black, S. (2004). Structural brain abnormalities in multiple sclerosis patients with major depression. Neurology, 62, 586–590. Fletcher, P. C., & Henson, R. N. (2001). Frontal lobes and human memory: Insights from functional neuroimaging. Brain, 124, 849–881. Gilboa, A., Ramirez, J., Kohler, S., Westmacott, R., Black, S. E., & Moscovitch, M. (2005). Retrieval of autobiographical memory in Alzheimer’s disease: Relation to volumes of medial temporal lobe and other structures. Hippocampus, 15, 535–550. Hodges, J. R., & Gurd, J. M. (1994). Remote memory and lexical retrieval in a case of frontal Pick’s disease. Archives of Neurology, 51, 821–827. Ivnik, R. J., Malec, J. F., Smith, G. E., Tangalos, E. G., & Peterson, R. C. (1996). Neuropsychological tests’ norms above age 55: COWAT, BNT, MAE token, WRAT-R Reading, AMNART, STROOP, TMT and JLO. The Clinical Neuropsychologist, 10, 262–278. Kapur, N. (1999). Syndromes of retrograde amnesia: A conceptual and empirical synthesis. Psychological Bulletin, 125, 800–825. Kertesz, A., Davidson, W., & Fox, H. (1997). Frontal behavioral inventory: Diagnostic criteria for frontal lobe dementia. Canadian Journal of Neurological Sciences, 24, 29–36. Kertesz, A., McMonagle, P., Blair, M., Davidson, W., & Munoz, D. G. (2005). The evolution and pathology of frontotemporal dementia. Brain, 128, 1996–2005. Kirwan, C. B., Bayley, P. J., Galvan, V. V., & Squire, L. R. (2008). Detailed recollection of remote autobiographical memory after damage to the medial temporal lobe. Proceedings of the National Academy of Sciences, U.S.A., 105, 2676–2680. Kopelman, M. D., Lasserson, D., Kingsley, D. R., Bello, F., Rush, C., Stanhope, N., et al. (2003). Retrograde amnesia and the volume of critical brain structures. Hippocampus, 13, 879–891. Kopelman, M. D., Wilson, B. A., & Baddeley, A. D. (1990). The autobiographical memory interview. Bury St. Edmunds: Thames Valley Test Company. Kovacevic, N., Henderson, J. T., Chan, E., Lifshitz, N., Bishop, J., Evans, A. C., et al. (2005). A three-dimensional MRI atlas of the mouse brain with estimates of the average and variability. Cerebral Cortex, 15, 639–645. Kovacevic, N., Lobaugh, N. J., Bronskill, M. J., Levine, B., Feinstein, A., & Black, S. E. (2002). A robust method for extraction and automatic segmentation of brain images. Neuroimage, 17, 1087–1100. Levine, B. (2004). Autobiographical memory and the self in time: Brain lesion effects, functional neuroanatomy, and lifespan development. Brain and Cognition, 55, 54–68. Levine, B., Kovacevic, N., Nica, I., Cheung, G., Schwartz, M. L., & Black, S. E. (2008). The Toronto traumatic brain injury study: Injury severity and quantified MRI. Neurology, 70, 771–778. Levine, B., Svoboda, E., Hay, J., Winocur, G., & Moscovitch, M. (2002). Aging and autobiographical memory: Dissociating episodic from semantic retrieval. Psychology and Aging, 17, 677–689. Matuszewski, V., Piolino, P., de la Sayette, V., Lalevee, C., Pelerin, A., Dupuy, B., et al. (2006). Retrieval mechanisms for autobiographical memories: Insights from the frontal variant of frontotemporal dementia. Neuropsychologia, 44, 2386–2397. McIntosh, A. R., Bookstein, F. L., Haxby, J. V., & Grady, C. L. (1996). Spatial pattern analysis of functional brain images using partial least squares. Neuroimage, 3, 143–157. McIntosh, A. R., Chau, W. K., & Protzner, A. B. (2004). Spatiotemporal analysis of event-related fMRI data using partial least squares. Neuroimage, 23, 764–775. McKhann, G. M., Albert, M. S., Grossman, M., Miller, B., Dickson, D., & Trojanowski, J. Q. (2001). Clinical and pathological diagnosis of frontotemporal dementia: Report of the Work Group on Frontotemporal Dementia and Pick’s Disease. Archives of Neurology, 58, 1803–1809. 1852 Journal of Cognitive Neuroscience Volume 20, Number 10 D o w n l o a d e d l l / / / / j f / t t i t . : / / f r o m D h o t w t n p o : a / d / e m d i t f r p o m r c h . s p i l d v i e r e r c c t h . m a i r e . d c u o o m c / n j a o r c t i n c / e a - p r d t i 2 c 0 l e 1 0 - p 1 d 8 f 3 / 9 2 1 0 9 / 3 1 6 0 6 / 1 1 0 8 o 3 c 9 n / 2 1 0 7 0 6 8 4 4 2 8 0 1 2 2 / 6 j o p c d n . b y 2 0 g 0 u 8 e . s t 2 o 0 n 1 2 0 6 7 . S p e d p f e m b y b e r g u 2 0 e 2 s 3 t / j f / . . . t o n 1 8 M a y 2 0 2 1 McKinnon, M. C., Black, S. E., Miller, B., Moscovitch, M., & Levine, B. (2006). Autobiographical memory in semantic dementia: Implication for theories of limbic–neocortical interaction in remote memory. Neuropsychologia, 44, 2421–2429. McKinnon, M. C., Svoboda, E., & Levine, B. (2007). The frontal lobes and autobiographical memory. In B. L. Miller & J. L. Cummings (Eds.), The human frontal lobes, functions and disorders (pp. 227–248). New York: Guilford Press. Miller, B. L., Seeley, W. W., Mychack, P., Rosen, H. J., Mena, I., & Boone, K. (2001). Neuroanatomy of the self: Evidence from patients with frontotemporal dementia. Neurology, 57, 817–821. Milner, B. (1963). Effects of different brain lesions on card sorting: The role of the frontal lobes. Archives of Neurology, 9, 100–110. Morris, R., Pandya, D. N., & Petrides, M. (1999). Fiber system linking the mid-dorsolateral frontal cortex with the retrosplenial/presubicular region in the rhesus monkey. Journal of Comparative Neurology, 407, 183–192. Morris, R., Petrides, M., & Pandya, D. N. (1999). Architecture and connections of retrosplenial area 30 in the rhesus monkey (Macaca mulatta). European Journal of Neuroscience, 11, 2506–2518. Moscovitch, M., Rosenbaum, R. S., Gilboa, A., Addis, D. R., Westmacott, R., Grady, C., et al. (2005). Functional neuroanatomy of remote episodic, semantic and spatial memory: A unified account based on multiple trace theory. Journal of Anatomy, 207, 35–66. Moss, H. E., Kopelman, M., Cappelletti, M., De Mornay Davies, P., & Jaldow, E. (2003). Lost for words or loss of memories? Autobiographical memory in semantic dementia. Cognitive Neuropsychology, 20, 703–732. Rubin, D. C., & Greenberg, D. L. (1998). Visual memory-deficit amnesia: A distinct amnesic presentation and etiology. Proceedings of the National Academy of Sciences, U.S.A., 95, 5413–5416. Rubin, D. C., & Schulkind, M. D. (1997). The distribution of autobiographical memories across the lifespan. Memory & Cognition, 25, 859–866. Simons, J. S., Graham, K. S., & Hodges, J. R. (2002). Perceptual and semantic contributions to episodic memory: Evidence from semantic dementia and Alzheimer’s disease. Journal of Memory and Language, 47, 197–213. Simons, J. S., & Spiers, H. J. (2003). Prefrontal and medial temporal lobe interactions in long-term memory. Nature Reviews Neuroscience, 4, 637–648. Simons, J. S., Verfaellie, M., Galton, C. J., Miller, B. L., Hodges, J. R., & Graham, K. S. (2002). Recollection-based memory in frontotemporal dementia: Implications for theories of long-term memory. Brain, 125, 2523–2536. Snowden, J., Griffiths, H., & Neary, D. (1994). Semantic dementia: Autobiographical contribution to preservation of meaning. Cognitive Neuropsychology, 11, 265–288. Snowden, J., Neary, D., & Mann, D. (2007). Frontotemporal lobar degeneration: Clinical and pathological relationships. Acta Neuropathologica (Berlin), 114, 31–38. So¨derlund, H., Black, S. E., Miller, B. L., Freedman, M., & Levine, B. (2008). Episodic memory and regional atrophy in frontotemporal lobar degeneration. Neuropsychologia, 46, 127–136. Spiers, H. J., Maguire, E. A., & Burgess, N. (2001). Hippocampal amnesia. Neurocase, 7, 357–382. Squire, L. R., & Bayley, P. J. (2007). The neuroscience of remote memory. Current Opinion in Neurobiology, 17, 185–196. Neary, D., Snowden, J. S., Gustafson, L., Passant, U., Stuss, St.-Jacques, P., & Levine, B. (2007). Aging and emotional D. T., Black, S., et al. (1998). Frontotemporal lobar degeneration: A consensus on clinical diagnostic criteria. Neurology, 51, 1546–1554. Nestor, P. J., Graham, K. S., Bozeat, S., Simons, J. S., & Hodges, J. R. (2002). Memory consolidation and the hippocampus: Further evidence from studies of autobiographical memory in semantic dementia and frontal variant frontotemporal dementia. Neuropsychologia, 40, 633–654. Piolino, P., Chetelat, G., Matuszewski, V., Landeau, B., Mezenge, F., Viader, F., et al. (2007). In search of autobiographical memories: A PET study in the frontal variant of frontotemporal dementia. Neuropsychologia, 45, 2730–2743. Piolino, P., Desgranges, B., Belliard, S., Matuszewski, V., Lalevee, C., De la Sayette, V., et al. (2003). Autobiographical memory and autonoetic consciousness: Triple dissociation in neurodegenerative diseases. Brain, 126, 2203–2219. Ratnavalli, E., Brayne, C., Dawson, K., & Hodges, J. R. (2002). The prevalence of frontotemporal dementia. Neurology, 58, 1615–1621. Rosen, H. J., Allison, S. C., Ogar, J. M., Amici, S., Rose, K., Dronkers, N., et al. (2006). Behavioral features in semantic dementia vs other forms of progressive aphasias. Neurology, 67, 1752–1756. Rosenbaum, R. S., McKinnon, M. C., Levine, B., & Moscovitch, M. (2004). Visual imagery deficits, impaired strategic retrieval, or memory loss: Disentangling the nature of an amnesic person’s autobiographical memory deficit. Neuropsychologia, 42, 1619–1635. Rosenbaum, R. S., Moscovitch, M., Foster, J. K., Verfaellie, M., Gao, F. Q., Black, S. E., et al. (2008). Patterns of autobiographical memory loss in medial temporal lobe amnesic patients. Journal of Cognitive Neuroscience, 20, 1490–1506. Rosene, D. L., & Van Hoesen, G. W. (1977). Hippocampal efferents reach widespread areas of cerebral cortex and amygdala in the rhesus monkey. Science, 198, 315–317. autobiographical memory. Memory, 15, 129–144. Steinvorth, S., Levine, B., & Corkin, S. (2005). Medial temporal lobe structures are needed to re-experience remote autobiographical memories: Evidence from H.M. and W.R. Neuropsychologia, 43, 479–496. Stuss, D. T., & Benson, D. F. (1986). The frontal lobes. New York: Raven Press. Svoboda, E., McKinnon, M. C., & Levine, B. (2006). The functional neuroanatomy of autobiographical memory: A meta-analysis. Neuropsychologia, 44, 2189–2208. Talairach, J., & Tournoux, P. (1988). Co-planar stereotaxic atlas of the human brain. Stuttgart: Georg Thieme Verlag. Tulving, E. (2002). Episodic memory: From mind to brain. Annual Review of Psychology, 53, 1–25. Valenstein, E., Bowers, D., Verfaellie, M., Heilman, K. M., Day, A., & Watson, R. T. (1987). Retrosplenial amnesia. Brain, 110, 1631–1646. Vogt, B. A., Pandya, D. N., & Rosene, D. L. (1987). Cingulate cortex of the rhesus monkey: I. Cytoarchitecture and thalamic afferents. Journal of Comparative Neurology, 262, 256–270. Wasserman, S., & Bockenholt, U. (1989). Bootstrapping: Applications to psychophysiology. Psychophysiology, 26, 208–221. Westmacott, R., & Moscovitch, M. (2003). The contribution of autobiographical significance to semantic memory. Memory & Cognition, 31, 761–774. Woods, R. P., Grafton, S. T., Holmes, C. J., Cherry, S. R., & Mazziotta, J. C. (1998). Automated image registration: I. General methods and intrasubject, intramodality validation. Journal of Computer Assisted Tomography, 22, 139–152. Woods, R. P., Grafton, S. T., Watson, J. D., Sicotte, N. L., & Mazziotta, J. C. (1998). Automated image registration: II. Intersubject validation of linear and nonlinear models. Journal of Computer Assisted Tomography, 22, 153–165. McKinnon et al. 1853 D o w n l o a d e d l l / / / / j t t f / i t . : / / f r o m D h o t w t n p o : a / d / e m d i t f r p o m r c h . s p i l d v i e r e r c c t h . m a i r e . d c u o o m c / n j a o r c t i n c / e a - p r d t i 2 c 0 l e 1 0 - p 1 d 8 f 3 / 9 2 1 0 9 / 3 1 6 0 6 / 1 1 0 8 o 3 c 9 n / 2 1 0 7 0 6 8 4 4 2 8 0 1 2 2 / 6 j o p c d n . b y 2 0 g 0 u 8 e . s t 2 o 0 n 1 2 0 6 7 . S p e d p f e m b y b e r g u 2 0 e 2 s 3 t / j . . t / . f o n 1 8 M a y 2 0 2 1

PDF Herunterladen