Abstract

Episodic memory has recently been shown to be impaired in the behavioral variant of frontotemporal dementia (bvFTD) when so-called non-progressive cases are excluded. Such non-progressive cases present with the behavioral features of bvFTD, but show no evidence of cognitive decline over time. To date, evidence regarding episodic memory performance in bvFTD subgroups on more stringent tasks is lacking. We investigated temporal and spatial source memory in progressive (n = 7) versus non-progressive (n = 12) bvFTD. BvFTD cases were retrospectively classified based on general cognitive decline on the Addenbrooke's Cognitive Examination Revised, and the presence of atrophy on structural neuroimaging, over 3 years following diagnosis. Progressors showed impaired temporal and spatial source retrieval. Non-progressors displayed temporal source deficits only. These differential source memory profiles point to the variability of episodic memory performance in bvFTD, and underscore the importance of differential diagnosis of bvFTD subgroups using longitudinal and neuroimaging data.

Introduction

Frontotemporal dementia (FTD) refers to a progressive neurodegenerative disorder characterized by atrophy in the frontal and/or temporal lobes, of which two broad presentations are recognized: a progressive deterioration in social function and personality known as behavioral variant FTD (bvFTD), and insidious decline in language skills referred to as primary progressive aphasia, which in turn can be subdivided into progressive non-fluent aphasia and semantic dementia (SD; Hodges et al., 2004; Hodges & Patterson, 2007; Neary, Snowden, Northen, & Goulding, 1988; Rascovsky et al., 2011). Patients diagnosed with bvFTD typically exhibit marked changes in behavior and personality (Piguet, Hornberger, Mioshi, & Hodges, 2011), whereas visuo-perceptual skills, semantic knowledge, and episodic memory appear intact.

Formal studies of episodic memory show considerable variation in memory performance in bvFTD (Glosser, Gallo, Clark, & Grossman, 2002; Kramer et al., 2005; Pasquier, Grymonprez, Lebert, & Van der Linden, 2001; Pennington, Hodges, & Hornberger, 2011; reviewed by Piguet et al., 2011). These inconsistencies across studies raise important questions regarding the accurate diagnosis of bvFTD, given that evidence of “severe amnesia” is considered as a diagnostic exclusion criterion for FTD (Neary et al., 1988; Rascovsky et al., 2011). Indeed, recent studies have demonstrated that bvFTD patients show episodic memory impairments of the same magnitude as patients with Alzheimer's disease across the standard neuropsychological memory tests (Pennington et al., 2011) and tests of autobiographical memory (Irish et al., 2011). Importantly, these profiles of episodic retreival could not be explained by differences in disease severity. Critically, Hornberger, Piguet, Graham, Nestor, and Hodges (2010) used postmortem data to show that a large percentage of bvFTD patients, with confirmed FTD pathology, present clinically with objective memory problems on neuropsychological tests.

The discovery of marked episodic memory deficits in bvFTD, dovetails with the identification of a subgroup of patients presenting with characteristic florid behavioral features suggestive of bvFTD, without actively progressing to frank dementia or showing evidence of structural atrophy on MRI scans (Kipps, Hodges, & Hornberger, 2010). Those non-progressive patients with scans in the control range were characterized by no, or very little, disease progression (Kipps et al., 2007) and had, therefore, significantly better disease prognosis (Davies et al., 2006), in contrast with patients with abnormal scans, who were largely institutionalized or dead within 3 years (Garcin et al., 2009). Importantly, the progressive subset of bvFTD patients have been shown to be as impaired as Alzheimer's disease patients on episodic memory tasks, whereas non-progressive bvFTD subgroups show subtle or patchy episodic memory impairments (Hornberger, Piguet, Kipps, & Hodges, 2008; Hornberger, Savage, et al., 2010; Pennington et al., 2011). The existence of such non-progressive cases, therefore, presents significant difficulties for the clinical diagnosis of bvFTD (Rascovsky et al., 2011) as these patients are indistinguishable from patients with progressive bvFTD when the original diagnostic criteria are applied (Kipps et al., 2007; Knibb, Kipps, & Hodges, 2006; Piguet, Hornberger, Shelley, Kipps, & Hodges, 2009).

To our knowledge, no study to date has directly contrasted progressive and non-progressive bvFTD patients on more stringent memory tasks that explicitly target the retrieval of contextual episodic information. One such task, commonly used in the memory field, is source memory, in which participants are required to encode and retrieve information about the context in which items were studied (Simons et al., 2002; Söderlund, Black, Miller, Freedman, & Levine, 2008; Vakil, Raz, & Levy, 2010). Source memory paradigms center on the idea that the contextual features of an event are bound together into a representation, and such specific details differentiate one event from another, making a memory episodic (Johnson, 2006; Mitchell & Johnson, 2009).

Mounting evidence suggests that dissociable brain regions may contribute to the retrieval of distinct source contextual details (Duarte, Henson, & Graham, 2011). Neuroimaging and patient studies have implicated the prefrontal cortex in supporting retrieval of temporal source information (Duarte, Henson, Knight, Emery, & Graham, 2010; Ekstrom, Copara, Isham, Wang, & Yonelinas, 2011; Fujii et al., 2004). In contrast, the medial temporal lobes, particularly the parahippocampal cortex, have been implicated in the successful retrieval of spatial source details (Ekstrom et al., 2011; Spiers & Maguire, 2007). In progressive bvFTD, the predominant pattern of atrophy is sited within the prefrontal cortex, with evidence of medial temporal lobe atrophy with disease progression (Kipps et al., 2007; Rosen et al., 2002; Seeley et al., 2008), providing a strong rationale for investigating source memory performance in this cohort.

The available evidence suggests an impaired capacity for bvFTD patients to recollect the encoding context of an event, including information about the set in which an item was presented (Simons et al., 2002) and the modality (visual vs. auditory) of presentation (Söderlund et al., 2008). In this study, we explored source memory accuracy in bvFTD using an experimental task designed to differentially stress temporal and spatial aspects of retrieval. We modified a temporal source memory task previously used in bvFTD (Simons et al., 2002) by adding a separate spatial source condition and matching the temporal and spatial source conditions for task difficulty. Given recent evidence concerning marked differences in disease progression within bvFTD cohorts, we were particularly interested to explore potential variations in source memory accuracy contingent on disease progression.

We hypothesized that global deficits in source memory would be evident only in progressive bvFTD cases, given their well-documented macroscopic atrophy in the prefrontal cortex (Rosen et al., 2002). In contrast, as there is no evidence that non-progressive cases show macroscopic atrophy on structural imaging, we predicted that these cases would score at control levels for both temporal and spatial source retrieval. To successfully demonstrate differential profiles of source memory performance in subgroups of bvFTD would provide further evidence for the fractionation of bvFTD cohorts into progressive and non-progressive subgroups based on their disease trajectory, reinforcing the importance of obtaining longitudinal clinical data as part of the diagnostic process in FTD syndromes.

Materials and Methods

Participants

We enrolled 19 patients with bvFTD and 16 healthy age-, sex-, and education-matched controls in this study. All participants gave informed consent to participate in the study, which was approved by the ethical committee of Addenbrooke's Hospital, Cambridge. All experimental work was carried out in accordance with the Declaration of Helsinki (1991).

All patients with bvFTD were recruited from the Early Dementia Clinics at Addenbrooke's Hospital, Cambridge. All patients met the original consensus criteria for FTD (Neary et al., 1988) with insidious onset, decline in social behavior and personal conduct, emotional blunting, and loss of insight (see Table 1 for demographic details). Patients were assessed by a multidisciplinary team to ensure the absence of other neurological or psychiatric (e.g., schizophrenia, depression, and mania) symptoms. All caregivers completed the Cambridge Behavioral Inventory (CBI; Bozeat, Gregory, Ralph, & Hodges, 2000; Wedderburn et al., 2008) to assess behavioral symptoms. The CBI has been validated against the Neuropsychiatric Inventory (Cummings, 1997) and shows good categorization of different dementias.

Table 1.

Demographic and clinical characteristics of bvFTD patients and controls (SD in parenthesis)

 Progressors Non-progressors Controls Progressors versus non-progressors 
N 12 16 — 
Mean age at test 63.0 (10.5) 61.5 (7.9) 62.0 .325 
Education 12.0 (2.4) 12.2 (2.1) 12.0 .451 
Sex (M/F) 6/1 11/1 14/2 .614 
NART premorbid IQ 113.3 (9.4) 107.4 (12.2) n/a .106 
ACE-R Total (100) 79.3 (11.8) 89.3 (6.7) 93.7 (4.3)a .022 
MMSE (30) 27.7 (2.9) 28.2 (2.0) 28.0 (1.7)b .418 
CBI (316) 108.4 (36) 76.8 (40.2) n/a .034 
 Progressors Non-progressors Controls Progressors versus non-progressors 
N 12 16 — 
Mean age at test 63.0 (10.5) 61.5 (7.9) 62.0 .325 
Education 12.0 (2.4) 12.2 (2.1) 12.0 .451 
Sex (M/F) 6/1 11/1 14/2 .614 
NART premorbid IQ 113.3 (9.4) 107.4 (12.2) n/a .106 
ACE-R Total (100) 79.3 (11.8) 89.3 (6.7) 93.7 (4.3)a .022 
MMSE (30) 27.7 (2.9) 28.2 (2.0) 28.0 (1.7)b .418 
CBI (316) 108.4 (36) 76.8 (40.2) n/a .034 

Notes: NART = National Adult Reading Test; ACE-R = Addenbrooke's Cognitive Examination Revised; MMSE = Mini-Mental State Examination; CBI = Cambridge Behavioral Inventory. Maximum test scores are shown in parenthesis.

General Cognitive Screening

All bvFTD patients were assessed across a battery of neuropsychological tests. Patients completed the Addenbrooke's Cognitive Examination Revised (ACE-R; Mioshi, Dawson, Mitchell, Arnold, & Hodges, 2006) and the Mini-Mental State Examination (MMSE; Folstein, Folstein & McHugh, 1975), as general indices of global cognitive functioning. The ACE-R covers five cognitive domains (Attention and Orientation, Memory, Fluency, Language, and Visuospatial). A maximum of 100 points can be achieved on the ACE-R, with scores below 88 suggestive of a dementia syndrome. The National Adult Reading Test (NART; Nelson & Willison, 1991) was administered as a measure of premorbid IQ. Further cognitive tests included verbal letter and category fluency (Spreen & Strauss, 1991), the Rey Complex Figure (RCF; Meyers & Meyers, 1995) as a measure of non-verbal episodic memory, the Rey Auditory Verbal Learning Task (RAVLT; Schmidt, 1996) as an index of verbal episodic memory, the backwards Digit span subtest from the Wechsler Memory Scale (Wechsler, 1998) and the Trail Making Task (Reitan, 1958) as measures of executive functioning. Patients also completed the Hayling sentence completion and the Brixton spatial anticipation tests (Burgess & Shallice, 1997), which measure response inhibition and perseveration, respectively. Finally, patients were assessed on the Visual Object and Space Perception battery (Warrington & James, 1991), which provides an index of visuospatial functioning covering two subscales of Dot Counting and Cube Counting.

Classification of FTD Patients

Over the following 3 years, all patients were seen at least once per year in the clinic for follow-up appointments and repeat cognitive testing. Two senior neurologists, blind to the subgroups, classified the cases retrospectively into progressive versus non-progressive cases. Classification of non-progressors was based on their lack of decline on the ACE-R general cognitive screening task over the 3-year period following diagnosis. This classification was made in line with the observations of a median survival rate of just 3 years following clinical presentation in bvFTD (Hodges, Davies, Xuereb, Kril, & Halliday, 2003).

Imaging data were also available for the majority of patients (n = 15), with atrophy on MRI at presentation used as supplementary information in the classification process, in line with the recently revised criteria for probable bvFTD (Rascovsky et al., 2011). One rater, a senior neurologist, blind to the clinical diagnosis, rated T1 coronal MRIs based on a validated rating scale (Kipps et al., 2007). This rater has been instrumental in designing and validating this reliable rating scale that involves reviewing two standardized coronal MRI slices; the first at the level of the temporal pole and the second at the level of the insula. Atrophy within each region was rated on a 5-point Likert scale ranging from 0 to 4 (0 = normal; 1 = borderline appearances, possibly normal; 2 = definite atrophy present; 3 = marked atrophy; 4 = severe atrophy). Non-progressive cases scored 0 or 1 across all brain regions on the visual MRI rating scale (Davies et al., 2006). Progressive cases scored from 1 to 3 across all brain regions. Importantly, there were no instances where a non-progressive case at diagnosis received an atrophy rating of >1 and were subsequently excluded from the study.

Comparison of the bvFTD subgroups' (seven progressive vs. eight non-progressive) MRI scans revealed significant differences for the following measured regions: Left frontal (U = 10.0, p = .020, r = .58), right frontal (U = 9.0, p = .014, r = .62), left posterior temporal (U = 10.0, p = .020, r = .60), and right posterior temporal (U = 11.5, p = .027, r = .58). No differences were present between the bvFTD subgroups for anterior temporal ratings (left: U = 22, p = .268, r = .22; right: U = 15.5, p = .076, r = .46).

Control participants were recruited through the Medical Research Council Cognition and Brain Sciences Unit (MRC-CBU) control panel. All controls were healthy with no neurological or psychiatric diseases. These participants were monitored over a period of years, and routinely screened for neurological and neuropsychological issues. Controls were paid £7.50 per hour for their participation in the study. No controls were excluded from this study.

Materials

The task used in this experiment was programmed in E-Prime version 1.1 SP3 (Psychology Software Tools, Inc., Pittsburgh, PA, USA), running on a laptop attached to a 15-inch LCD touchscreen. The screen was positioned approximately 80 cm from the subject and responses collected using an analog response box (Psychology Software Tools, Inc.). In constructing the task, four sets of 30 line drawings (named A, B, C, & D) were selected from the Snodgrass and Vanderwart (1980) corpus. These sets were matched for ratings of concept familiarity and for the proportions of living/non-living items. Four versions of the task were then prepared, which counterbalanced the four stimuli sets as well as the condition order (temporal/spatial vs. spatial/temporal). Patients were tested on the experimental measures during their initial visit to the clinic.

Experimental Design and Procedure

A modified version of the source monitoring task described by Simons and colleagues (2002) and Duarte and colleagues (2010) was used, involving two study and two test phases (Figure 1). In each of the study phases, from now on referred to as Sets 1 and 2, participants saw 30 line drawings presented for 15 s each. Subjects were asked to name and provide as much semantic information about the depicted item as possible to ensure a deep level of encoding. Incidental to this task, the stimuli were shown either in the upper half or the lower half of the screen during the study phases. The vertical displacement of drawings varied in that some stimuli were presented close to the horizontal midline of the screen, whereas others were presented more distant from the midline. This manipulation was employed, following extensive piloting in healthy participants, to equate spatial and temporal source performance in controls by increasing the difficulty of the spatial location decision. Paired-samples t-tests revealed no significant differences between temporal and spatial source accuracy (t = 0.310, p = .761) or between temporal and spatial reaction times (t = 1.279, p = .220) in controls, indicating the two experimental conditions were matched for the level of difficulty. Sets 1 and 2 of the study phase were separated by a filled delayed period of 10 min during which unrelated tasks were performed by the participants. The delay between Sets 1 and 2 ensured a sufficient temporal distinction between both sets for the temporal source judgments.

Fig. 1.

Diagram showing the procedure for Test Phases for Temporal and Spatial conditions.

Fig. 1.

Diagram showing the procedure for Test Phases for Temporal and Spatial conditions.

At test, all 60 drawings from Sets 1 and 2 were viewed again, randomly interspersed with 60 novel foil drawings, and subjects were asked to make, separately, spatial and temporal source judgments. In the temporal condition, 30 previously studied drawings (15 from Set 1 and 15 from Set 2; of which 15 had been seen in the upper half of the screen and 15 had been seen in the lower half of the screen) were re-presented mixed up with 30 novel foil drawings. For each item, participants had to decide whether they had seen the item in Set 1, Set 2, or not at all. In the spatial condition, the remaining 30 studied line drawings together with 30 further novel foil drawings were shown. In this condition, participants were required to decide whether they had seen each item in the upper or lower half of the screen, or not at all. Subjects responded via a serial response box and key assignments were kept constant across both test conditions and all participants. The test phase responses were self-paced, with a prompt by the experimenter after 15 s to respond, when necessary.

Source Memory Scoring

Response frequencies for the spatial and temporal conditions can be found in Table 3. Recognition memory accuracy was defined as the proportion of items reported as having been seen before (hits), regardless of correct source, minus the proportion of novel foil items reported as having been seen before (false alarms). Source memory accuracy was defined as the proportion of previously studied items correctly attributed to their source, minus the proportion of previously studied items incorrectly attributed to their source.

Statistical Analyses

Data were analyzed using SPSS 15.0 (SPSS Inc., Chicago, IL, USA). A priori, all variables were plotted and checked for normality of distribution using the Kolmogorov–Smirnov tests. Most variables revealed non-normal distributions. Given the small sample sizes under consideration, and non-normal distribution of variables, non-parametric statistical tests were therefore used. Mann–Whitney U-tests were employed to explore differences across demographic (age, education), neuropsychological (memory, executive function, language, visuospatial and general cognitive tests), atrophy ratings, and experimental source memory data between the three groups (controls, progressive bvFTD, and non-progressive bvFTD). Estimates of effect size were calculated for all comparisons (Rosenthal, 1991).

Results

Demographic and Clinical Characteristics

Participants were matched for age, years in education, and male:female ratio (Table 1). Progressors scored significantly lower than age-matched norms on the ACE-R measure of general cognitive function; however, non-progressors scored within normal levels. No differences were evident across groups for total score on the MMSE. Comparisons of progressive versus non-progressive cases revealed no significant difference across any of the demographic variables (Table 1). Progressive and non-progressive cases did not differ significantly from each other for premorbid IQ on the NART (p = .106) or general cognitive functioning on the MMSE (p = .418). Progressive bvFTD cases scored significantly lower than non-progressive cases on the ACE-R total score (U = 18.0, p = .022; r = .47), and the memory (U = 18.0, p = .022, r = .47) and verbal fluency (U = 20.5, p = .034, r = .42) subscales of the ACE-R. Progressive cases also received higher ratings of behavioral dysfunction by their carers on the CBI (U = 20.0, p = .034, r = .43).

Cognitive Profiles of bvFTD Subgroups

Progressive bvFTD patients showed significant impairments in episodic memory, scoring below 2 SD in comparison with published normative scores on the RAVLT (immediate recall, delayed recall, delayed recognition). Progressive cases were also impaired for Category Fluency, the Brixton spatial anticipation task, and the Trail Making Test, indices of executive function. Non-progressive bvFTD cases scored within normal levels for all neuropsychological tests except the Trail Making Test, where they performed slower than published normative scores.

Comparing the progressive and non-progressive cases revealed largely similar cognitive performance, with no significant differences evident across the majority of tests of executive function (Table 2). In contrast, the FTD subgroups differed significantly on tests of episodic memory, with progressive cases scoring significantly lower than non-progressive cases for delayed recall of the RCF (U = 19.0, p = .028, r = .45), and immediate recall (U = 6.0, p = .006, r = .57), delayed recall (U = 6.0, p = .006, r = .57), and recognition (U = 5.0, p = .004, r = .61) subscales of the RAVLT.

Table 2.

Background neuropsychological data for bvFTD patients and normative control data (SD in parenthesis)

Neuropsychological tests Progressors Non-progressors Control normative data Progressors versus non-progressors Effect size (r
Memory 
 Rey Complex Figure: Recall 8.5 (6.1) 16.6 (9.2) 15.9 (5.9)a .028 .45 
 RAVLT: Immediate recall 3.0 (1.8) 7.4 (3.0) 9.3 (2.9)b .006 .57 
 RAVLT: 30 min recall 1.8 (1.9) 6.1 (2.8) 8.8 (3.0)b .006 .57 
 RAVLT: Recognition 9.0 (3.3) 13.4 (1.9) 13.5 (1.3)b .004 .61 
Executive function 
 Digit span: Backward 4.6 (2.2) 4.7 (1.4) 7.3 (1.8)c .296 .14 
 Trails B-A (s) 82.0 (67.7) 79.6 (49.6) 41.3 (10.4)d .449 .15 
 Hayling: Scaled score 4.1 (1.9) 5.5 (1.1) 6.4 (1.7)e .083 .36 
 Brixton: Scaled score 4.3 (1.0) 6.3 (1.8) 6.0e .008 .56 
 Letter fluency: Total correct 22.4 (21.2) 31.5 (16.8) 32.3 (12.7)f .075 .33 
 Category fluency: Total 23.1 (14.8) 43.9 (13.9) 45.2 (9.6)g .004 .59 
Visuospatial 
 Rey Complex Figure: Copy 33.1 (3.4) 35.1 (1.1) 30.8 (4.2)a .075 .36 
 VOSP: Dot counting 10.0 (0) 9.8 (0.4) 9.8 (0.6)h .296 .25 
 VOSP: Cube analysis 9.9 (.4) 9.2 (1.4) 9.5 (0.8)h .241 .23 
Neuropsychological tests Progressors Non-progressors Control normative data Progressors versus non-progressors Effect size (r
Memory 
 Rey Complex Figure: Recall 8.5 (6.1) 16.6 (9.2) 15.9 (5.9)a .028 .45 
 RAVLT: Immediate recall 3.0 (1.8) 7.4 (3.0) 9.3 (2.9)b .006 .57 
 RAVLT: 30 min recall 1.8 (1.9) 6.1 (2.8) 8.8 (3.0)b .006 .57 
 RAVLT: Recognition 9.0 (3.3) 13.4 (1.9) 13.5 (1.3)b .004 .61 
Executive function 
 Digit span: Backward 4.6 (2.2) 4.7 (1.4) 7.3 (1.8)c .296 .14 
 Trails B-A (s) 82.0 (67.7) 79.6 (49.6) 41.3 (10.4)d .449 .15 
 Hayling: Scaled score 4.1 (1.9) 5.5 (1.1) 6.4 (1.7)e .083 .36 
 Brixton: Scaled score 4.3 (1.0) 6.3 (1.8) 6.0e .008 .56 
 Letter fluency: Total correct 22.4 (21.2) 31.5 (16.8) 32.3 (12.7)f .075 .33 
 Category fluency: Total 23.1 (14.8) 43.9 (13.9) 45.2 (9.6)g .004 .59 
Visuospatial 
 Rey Complex Figure: Copy 33.1 (3.4) 35.1 (1.1) 30.8 (4.2)a .075 .36 
 VOSP: Dot counting 10.0 (0) 9.8 (0.4) 9.8 (0.6)h .296 .25 
 VOSP: Cube analysis 9.9 (.4) 9.2 (1.4) 9.5 (0.8)h .241 .23 

Notes: RAVLT = Rey Auditory Verbal Learning Task; VOSP = Visual Object and Space Perception. All scores presented for bvFTD patients are raw scores, unless otherwise stated. Effect sizes refer to progressive versus non-progressive bvFTD contrast. Normative scores refer to published norms from the following sources.

Source Memory Performance

First, we considered the source memory performance of the entire bvFTD group, irrespective of progression (n = 19) in comparison with controls (Figure 2). BvFTD cases performed significantly poorer than controls for overall recognition accuracy (U = 72.0, p = .003, r = .46) and for overall source accuracy (U = 63.0, p = .001, r = .50). For temporal source accuracy the combined bvFTD group was impaired with respect to controls (U = 47.0, p < .0001, r = .59), whereas no significant differences were evident for spatial source accuracy (U = 115.0, p = .115, r = .21).

Fig. 2.

Boxplots showing overall recognition and source accuracy (% correct) for progressive and non-progressive bvFTD patients and control participants. Whiskers indicate 5–95 percentile.

Fig. 2.

Boxplots showing overall recognition and source accuracy (% correct) for progressive and non-progressive bvFTD patients and control participants. Whiskers indicate 5–95 percentile.

Source Accuracy in Progressive bvFTD

The source accuracy performance of the progressive and non-progressive bvFTD subgroups is shown in Table 3. Progressive bvFTD cases were significantly impaired for overall recognition accuracy (U = 23.5, p = .013, r = .38) and overall source accuracy (U = 14, p = .002, r = .47) in comparison with controls. Further, both temporal source accuracy (U = 13.0, p = .001, r = .49) and spatial source accuracy (U = 30.0, p = .044, r = .30) was significantly compromised in the progressive bvFTD subgroup.

Table 3.

Source memory performance for bvFTD patients and controls (SD in parenthesis)

Conditions (% correct) bvFTD combined Progressors Non-progressors Controls Progressors versus non-progressors (p-value) Effect size (r
Overall recognition 94.6 (4.7) 93.4 (6.0) 95.3 (3.9) 98.2 (2.6) .319 .11 
Overall source 66.8 (11.8) 62.1 (11.3) 69.4 (11.7) 77.9 (7.4) .071 .34 
Temporal source 62.3 (13.2) 56.3 (15.2) 65.8 (11.1) 78.4 (9.1) .028 .45 
Spatial source 71.3 (15.4) 68.0 (12.7) 73.3 (17.0) 77.4 (10.5) .355 .10 
Conditions (% correct) bvFTD combined Progressors Non-progressors Controls Progressors versus non-progressors (p-value) Effect size (r
Overall recognition 94.6 (4.7) 93.4 (6.0) 95.3 (3.9) 98.2 (2.6) .319 .11 
Overall source 66.8 (11.8) 62.1 (11.3) 69.4 (11.7) 77.9 (7.4) .071 .34 
Temporal source 62.3 (13.2) 56.3 (15.2) 65.8 (11.1) 78.4 (9.1) .028 .45 
Spatial source 71.3 (15.4) 68.0 (12.7) 73.3 (17.0) 77.4 (10.5) .355 .10 

Notes: bvFTD = behavioral variant frontotemporal dementia. Effect size refers to the progressive versus non-progressive subgroup comparisons.

Source Accuracy in Non-progressive bvFTD

Non-progressive bvFTD cases showed significant impairments for overall recognition accuracy (U = 48.5, p = .013, r = .38) and overall source accuracy (U = 49.0, p = .014, r = .37) in comparison with controls. Interestingly, although the non-progressive cases showed significant temporal source memory impairments (U = 34.0, p = .001, r = .49), no deficits were evident on the spatial source memory task (U = 85.0, p = .315, r = .09).

Progressive Versus Non-progressive bvFTD Source Memory Performance

Direct comparison of the bvFTD subgroups on the source memory tasks showed that progressive and non-progressive cases did not differ significantly for overall recognition accuracy (U = 36.5, p = .319, r = .11) or overall source accuracy (U = 24.5, p = .071, r = .34). Progressive bvFTD cases showed significantly poorer temporal source accuracy in comparison with non-progressors (U = 19.0, p = .028, r = .45). No significant differences were evident between the bvFTD subgroups on the spatial source accuracy task (U = 37.0, p = .319, r = .11).

Discussion

Our results reveal differential patterns of source memory impairment in progressive versus non-progressive subtypes of bvFTD. When the bvFTD group was considered as a whole, deficits were evident only on the temporal source task. Following a subgroup analysis, progressive bvFTD patients showed impaired source memory irrespective of condition with significant deficits across the temporal and spatial source tasks. In contrast, non-progressive cases displayed a selective impairment of temporal source memory, which was less pronounced than that observed in the progressive subgroup. Here, we discuss the implications of our findings for understanding episodic memory dysfunction in bvFTD.

Our findings of disrupted source memory in progressive bvFTD are consistent with previous studies (Simons et al., 2002; Söderlund et al., 2008). Such source memory difficulties in encoding the context of an event are to be expected in progressive bvFTD given the link between frontal lobe function and source memory performance (Duarte et al., 2010; Glisky, Polster, & Routhieaux, 1995). Importantly, we have demonstrated that bvFTD patients were impaired not only for temporal order memory, but also for spatial source retrieval. The disruptions of temporal order and spatial source memory in progressive bvFTD may, in part, account for the recent findings of impaired episodic memory in progressive bvFTD on the standard neuropsychological tests (Hornberger, Piguet, et al., 2010) in particular for recall-based neuropsychological tests which stress more temporal order processes. Whether previous inconsistencies in episodic memory performance in bvFTD reflect the mixture of progressive and non-progressive subgroups of bvFTD or are attributable to other variables such as task differences, remains unknown. Deficits in source memory processes might also account for autobiographical memory impairments in bvFTD (Irish et al., 2011). To lose the “where” and “when” of recent and remote episodes renders recollection of such events essentially meaningless, when we consider that while item information can reoccur across different contexts, it is the source information that ascribes an event its unique signature (Ekstrom et al., 2011).

Of note is the finding that source memory deficits were also evident in the non-progressive bvFTD subgroup, albeit exclusively in the temporal source condition. These temporal source deficits are somewhat surprising in light of their lack of frontal and/or temporal atrophy on MRI (Kipps et al., 2010; Pennington et al., 2011) and their relatively intact performance on neuropsychological memory tests. Interestingly, temporal source difficulties in non-progressive bvFTD were at an intermediate level to those of progressive cases. This corroborates previous studies which have shown that although non-progressive patients perform significantly better than progressive bvFTD patients, their performance remains lower than that of healthy controls (Hornberger, Piguet, et al., 2010), thus adding to the diagnostic uncertainty of this cohort. Closer inspection of the source memory data revealed consistent performance within the non-progressive group for temporal and spatial source accuracy, with the exception of one individual who achieved 93% accuracy on the temporal source task. Critically, this individual also scored within control levels on the Trail Making Task, suggesting that executive function and frontally mediated processes may have contributed to his successful source memory performance. In contrast, the remaining patients in the non-progressive group performed poorly on the Trail Making Test, which might point to some form of prefrontal cortex dysfunction, potentially on a microscopic or neurotransmitter level. Accordingly, the disruption of prefrontal cortex-mediated processes, such as temporal source memory, would therefore be expected in this group, while temporal lobe-mediated processes, such as spatial source memory, should remain intact, as observed. Nevertheless, this dissociation between temporal and spatial source memory in bvFTD subgroups, and its underlying mechanisms, warrants further investigation.

The unclear etiology of non-progressive bvFTD cases has led to recent debate concerning their status. To date, no neuropathological confirmation of actual FTD pathology in non-progressive bvFTD cases has been established, and this represents a critical line of enquiry to establish whether their lack of atrophy on MRI represents a benign disease trajectory (Koedam et al., 2010) or if they warrant consideration as a separate cohort with subclinical psychiatric symptoms (Hornberger et al., 2008). One proposition is that a proportion of non-progressive bvFTD cases may harbor long-standing personality disorders that lie along the Asperger spectrum and present with a decompensation later in life (Davies et al., 2006; Hornberger et al., 2008). To further explore the etiology of non-progressive patients, it would be interesting to compare non-progressive bvFTD cases with psychiatric disorders that can present with behavioral characteristics similar to bvFTD. These conditions include bipolar disorder in which changes in the prefrontal cortex are evident (Strakowski, Delbello, & Adler, 2005), and schizophrenia in which source memory deficits have been documented (Brébion, Gorman, Amador, Malaspina, & Sharif, 2002).

A number of methodological limitations warrant consideration. First, we were limited to a relatively small number of bvFTD cases in which there is a preponderance of non-progressors. This may reflect the compliance of non-progressors to undergo extensive cognitive testing, in comparison with progressive bvFTD cases in which apathy and lack of motivation are the most common initial symptoms (Chow et al., 2009; Rascovsky et al., 2011). It is not possible from our study to determine whether impairments in temporal source memory in the non-progressive bvFTD subgroup reflect microscopic structural deficits (not obvious on MRI) or functional disturbance of the frontotemporal circuitry (e.g., neurotransmitter imbalance). Given that we used ratings of atrophy to divide our bvFTD cohort into subgroups, it will be important to replicate this study using a larger sample of patients who have all undergone structural neuroimaging. This would also offer the opportunity to investigate possible differences in the neural correlates of source memory dysfunction in the two bvFTD cohorts.

Our findings raise interesting future avenues for exploration, and underscore the ongoing challenge in refining the differential diagnosis of progressive versus non-progressive bvFTD cases. This represents a critical area of investigation given that survival rates and disease prognosis vary widely contingent on the bvFTD subgroup (Davies et al., 2006). To elucidate the precise variant of bvFTD at the point of diagnosis is likely to impact dramatically on the degree of stress and uncertainty experienced by patients and their family members. Indeed, knowledge of the potential disease trajectory and prognosis at the time of diagnosis would benefit carers in determining appropriate supports and behavioral interventions for the patient. Ultimately, the gold standard for the classification of bvFTD remains at postmortem, however, converging evidence regarding differential impairments on measures of episodic memory may offer a potentially useful means for assisting in the prospective identification of individuals most likely in the progressive subgroup.

In summary, we have demonstrated differential impairments across subgroups of bvFTD at the time of diagnosis using a novel measure of temporal and spatial source memory. These differential profiles of source memory impairment add to a growing body of evidence for the dissociation between progressive and non-progressive subtypes of bvFTD. Future research is needed to explore the underlying mechanisms resulting in divergent patterns of cognitive functioning in subgroups of bvFTD as well as establishing the etiology of the non-progressive subgroup using neuropsychological, neuroimaging, and pathological data.

Funding

This work was supported by a Wellcome Fellowship to AG, a Medical Research Council programme grant to JRH, Medical Research Council core funding to KSG, and an Alzheimer's Research Trust Grant to KSG and JRH.

Conflict of Interest

None declared.

Acknowledgements

The authors are grateful to the participants who gave their time to this study. The authors wish to thank Dr. Christopher Kipps for the MRI scan ratings.

References

Acevedo
A.
Loewenstein
D. A.
Barker
W. W.
Harwood
D. G.
Luis
C.
Bravo
M.
, et al.  . 
Category Fluency Test: Normative data for English-and Spanish-speaking elderly
Journal of the International Neuropsychological Society
 , 
2000
, vol. 
6
 (pg. 
760
-
769
)
Bonello
P. J.
Rapport
L. J.
Millis
S. R.
Psychometric properties of the visual object and space perception battery in normal older adults
The Clinical Neuropsychologist
 , 
1997
, vol. 
11
 (pg. 
436
-
442
)
Bozeat
S.
Gregory
C. A.
Ralph
M. A.
Hodges
J. R.
Which neuropsychiatric and behavioural features distinguish frontal and temporal variants of frontotemporal dementia from Alzheimer's disease?
Journal of Neurology Neurosurgery and Psychiatry
 , 
2000
, vol. 
69
 (pg. 
178
-
186
)
Brébion
G.
Gorman
J. M.
Amador
X.
Malaspina
D.
Sharif
Z.
Source monitoring impairments in schizophrenia: Characterisation and associations with positive and negative symptomatology
Psychiatry Research
 , 
2002
, vol. 
112
 (pg. 
27
-
39
)
Burgess
P.
Shallice
T.
The Hayling and Brixton Tests
 , 
1997
Thurston Suffolk
Thames Valley Test Company
Chow
T. W.
Binns
M. A.
Cummings
J. L.
Lam
I.
Black
S. E.
Miller
B. L.
, et al.  . 
Apathy symptom profile and behavioral associations in frontotemporal dementia vs dementia of Alzheimer type
Archives of Neurology
 , 
2009
, vol. 
66
 (pg. 
888
-
893
)
Cummings
J. L.
The Neuropsychiatric Inventory: Assessing psychopathology in dementia patients
Neurology
 , 
1997
, vol. 
48
 (pg. 
S10
-
S16
)
Davies
R. R.
Kipps
C. M.
Mitchell
J.
Kril
J. J.
Halliday
G. M.
Hodges
J. R.
Progression in frontotemporal dementia: Identifying a benign behavioral variant by magnetic resonance imaging
Archives of Neurology
 , 
2006
, vol. 
63
 (pg. 
1627
-
1631
)
Duarte
A.
Henson
R. N.
Graham
K. S.
Stimulus content and the neural correlates of source memory
Brain Research
 , 
2011
, vol. 
1373
 (pg. 
110
-
123
)
Duarte
A.
Henson
R. N.
Knight
R. T.
Emery
T.
Graham
K. S.
Orbito-frontal cortex is necessary for temporal context memory
Journal of Cognitive Neuroscience
 , 
2010
, vol. 
22
 (pg. 
1819
-
1831
)
Ekstrom
A. D.
Copara
M. S.
Isham
E. A.
Wang
W. C.
Yonelinas
A. P.
Dissociable networks involved in spatial and temporal order source retrieval
Neuroimage
 , 
2011
, vol. 
56
 (pg. 
1803
-
1813
)
Fastenau
P. S.
Denburg
N. L.
Hufford
B. J.
Adult norms for the Rey-Osterrieth Complex Figure Test and for supplemental recognition and matching trials from the Extended Complex Figure Test
The Clinical Neuropsychologist
 , 
1999
, vol. 
13
 (pg. 
30
-
47
)
Folstein
M. F.
Folstein
S. E.
McHugh
P. R.
“Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician
Journal of Psychiatry Research
 , 
1975
, vol. 
12
 (pg. 
189
-
198
)
Fujii
T.
Suzuki
M.
Okuda
J.
Ohtake
H.
Tanji
K.
Yamaguchi
K.
, et al.  . 
Neural correlates of context memory with real-world events
Neuroimage
 , 
2004
, vol. 
21
 (pg. 
1596
-
1603
)
Garcin
B.
Lillo
P.
Hornberger
M.
Piguet
O.
Dawson
K.
Nestor
P. J.
, et al.  . 
Determinants of survival in behavioral variant frontotemporal dementia
Neurology
 , 
2009
, vol. 
73
 (pg. 
1656
-
1661
)
Glisky
E. L.
Polster
M. R.
Routhieaux
B. C.
Double dissociation between item and source memory
Neuropsychology
 , 
1995
, vol. 
9
 (pg. 
229
-
235
)
Glosser
G.
Gallo
J. L.
Clark
C. M.
Grossman
M.
Memory encoding and retrieval in frontotemporal dementia and Alzheimer's disease
Neuropsychology
 , 
2002
, vol. 
16
 (pg. 
190
-
196
)
Hodges
J. R.
Davies
R. R.
Xuereb
J. H.
Casey
B.
Broe
M.
Bak
T. H.
, et al.  . 
Clinicopathological correlates in frontotemporal dementia
Annals of Neurology
 , 
2004
, vol. 
56
 (pg. 
399
-
406
)
Hodges
J. R.
Davies
R. R.
Xuereb
J. H.
Kril
J.
Halliday
G.
Survival in frontotemporal dementia
Neurology
 , 
2003
, vol. 
61
 (pg. 
349
-
354
)
Hodges
J. R.
Patterson
K.
Semantic dementia: A unique clinicopathological syndrome
The Lancet Neurology
 , 
2007
, vol. 
6
 (pg. 
1004
-
1014
)
Hornberger
M.
Piguet
O.
Graham
A. J.
Nestor
P. J.
Hodges
J. R.
How preserved is episodic memory in behavioral variant frontotemporal dementia?
Neurology
 , 
2010
, vol. 
74
 (pg. 
472
-
479
)
Hornberger
M.
Piguet
O.
Kipps
C. M.
Hodges
J. R.
Executive function in progressive and nonprogressive behavioral variant frontotemporal dementia
Neurology
 , 
2008
, vol. 
71
 (pg. 
1481
-
1488
)
Hornberger
M.
Savage
S.
Hsieh
S.
Mioshi
E.
Piguet
O.
Hodges
J. R.
Orbitofrontal dysfunction discriminates behavioral variant frontotemporal dementia from Alzheimer's disease
Dementia and Geriatric Cognitive Disorders
 , 
2010
, vol. 
30
 (pg. 
547
-
552
)
Irish
M.
Hornberger
M.
Lah
S.
Miller
L.
Pengas
G.
Nestor
P. J.
, et al.  . 
Profiles of recent autobiographical memory retrieval in semantic dementia, behavioural-variant frontotemporal dementia, and Alzheimer's disease
Neuropsychologia
 , 
2011
, vol. 
49
 (pg. 
2694
-
2702
)
Jacobson
M. W.
Delis
D. C.
Bondi
M. W.
Salmon
D. P.
Asymmetry in auditory and spatial attention span in normal elderly genetically at risk for Alzheimer's disease
Journal of Clinical and Experimental Neuropsychology
 , 
2005
, vol. 
27
 (pg. 
240
-
253
)
Johnson
M.
Memory and reality
The American Psychologist
 , 
2006
, vol. 
61
 (pg. 
760
-
771
)
Kipps
C. M.
Davies
R. R.
Mitchell
J.
Kril
J. J.
Halliday
G. M.
Hodges
J. R.
Clinical significance of lobar atrophy in frontotemporal dementia: Application of an MRI Visual Rating Scale
Dementia and Geriatric Cognitive Disorders
 , 
2007
, vol. 
23
 (pg. 
334
-
342
)
Kipps
C. M.
Hodges
J. R.
Hornberger
M.
Nonprogressive behavioural frontotemporal dementia: Recent developments and clinical implications of the “bvFTD phenocopy syndrome”
Current Opinion in Neurology
 , 
2010
, vol. 
23
 (pg. 
628
-
632
)
Knibb
J. A.
Kipps
C. M.
Hodges
J. R.
Frontotemporal dementia
Current Opinion in Neurology
 , 
2006
, vol. 
19
 (pg. 
565
-
571
)
Koedam
E. L. G. E.
Van der Flier
W. M.
Barkhof
F.
Koene
T.
Scheltens
P.
Pijnenburg
Y. A. L.
Clinical characteristics of patients with frontotemporal dementia with and without lobar atrophy on MRI
Alzheimer Disease and Associated Disorders
 , 
2010
, vol. 
24
 (pg. 
242
-
247
)
Kramer
J. H.
Rosen
H. J.
Du
A. T.
Schuff
N.
Hollnagel
C.
Weiner
M. W.
, et al.  . 
Dissociations in hippocampal and frontal contributions to episodic memory performance
Neuropsychology
 , 
2005
, vol. 
19
 (pg. 
799
-
805
)
Loonstra
A. S.
Tarlow
A. R.
Sellers
A. H.
COWAT metanorms across age, education, and gender
Applied Neuropsychology
 , 
2001
, vol. 
8
 (pg. 
161
-
166
)
Meyers
J.
Meyers
K.
The Meyers scoring system for the Rey complex figure and the recognition trial: Professional manual
 , 
1995
Odessa, FL
Psychological Assessment Resources
Mioshi
E.
Dawson
K.
Mitchell
J.
Arnold
R.
Hodges
J. R.
The Addenbrooke's Cognitive Examination Revised (ACE R): A brief cognitive test battery for dementia screening
International Journal of Geriatric Psychiatry
 , 
2006
, vol. 
21
 (pg. 
1078
-
1085
)
Mitchell
K. J.
Johnson
M. K.
Source monitoring 15 years later: What have we learned from fMRI about the neural mechanisms of source memory?
Psychological Bulletin
 , 
2009
, vol. 
135
 (pg. 
638
-
677
)
Neary
D.
Snowden
J. S.
Northen
B.
Goulding
P.
Dementia of frontal lobe type
Journal of Neurology, Neurosurgery and Psychiatry
 , 
1988
, vol. 
51
 (pg. 
353
-
361
)
Nelson
H. E.
Willison
J. R.
The Revised National Adult Reading Test—Test Manual
 , 
1991
Windsor
NFER-Nelson
Pasquier
F.
Grymonprez
L.
Lebert
F.
Van der Linden
M.
Memory impairment differs in frontotemporal dementia and Alzhemier's disease
Neurocase
 , 
2001
, vol. 
7
 (pg. 
161
-
171
)
Pennington
C.
Hodges
J. R.
Hornberger
M.
Neural correlates of episodic memory in behavioural variant frontotemporal dementia
Journal of Alzheimer's Disease
 , 
2011
, vol. 
24
 (pg. 
261
-
268
)
Piguet
O.
Hornberger
M.
Mioshi
E.
Hodges
J. R.
Behavioural-variant frontotemporal dementia: Diagnosis, clinical staging, and management
The Lancet Neurology
 , 
2011
, vol. 
10
 (pg. 
162
-
172
)
Piguet
O.
Hornberger
M.
Shelley
B. P.
Kipps
C. M.
Hodges
J. R.
Sensitivity of current criteria for the diagnosis of behavioral variant frontotemporal dementia
Neurology
 , 
2009
, vol. 
72
 (pg. 
732
-
737
)
Rascovsky
K.
Hodges
J. R.
Knopman
D.
Mendez
M. F.
Kramer
J. H.
Neuhaus
J.
, et al.  . 
Sensitivity of revised diagnostic criteria for the behavioural variant of frontotemporal dementia
Brain
 , 
2011
, vol. 
134
 (pg. 
2456
-
2477
)
Reitan
R.
Validity of the Trail Making Test as an indicator of organic brain damage
Perceptual and Motor Skills
 , 
1958
, vol. 
8
 (pg. 
271
-
276
)
Rosen
H. J.
Gorno-Tempini
M. L.
Goldman
W. P.
Perry
R. J.
Schuff
N.
Weiner
M.
, et al.  . 
Patterns of brain atrophy in frontotemporal dementia and semantic dementia
Neurology
 , 
2002
, vol. 
58
 (pg. 
198
-
208
)
Rosenthal
R.
Meta-analytic procedures for social research
 , 
1991
2nd ed.
Newbury Park, CA
Sage
Schmidt
M.
Rey Auditory and Verbal Learning Test: A handbook
 , 
1996
Los Angeles
Western Psychological Services
Seeley
W. W.
Crawford
R.
Rascovsky
K.
Kramer
J. H.
Weiner
M.
Miller
B. L.
, et al.  . 
Frontal paralimbic network atrophy in very mild behavioral variant frontotemporal dementia
Archives of Neurology
 , 
2008
, vol. 
65
 (pg. 
249
-
255
)
Simons
J. S.
Verfaellie
M.
Galton
C. J.
Miller
B. L.
Hodges
J. R.
Graham
K. S.
Recollection-based memory in frontotemporal dementia: Implications for theories of long-term memory
Brain
 , 
2002
, vol. 
125
 (pg. 
2523
-
2536
)
Snodgrass
J. G.
Vanderwart
M.
A standardized set of 260 pictures: Norms for name agreement, image agreement, familiarity, and visual complexity
Journal of Experimental Psychology: Human Learning and Memory
 , 
1980
, vol. 
6
 (pg. 
174
-
215
)
Söderlund
H.
Black
S. E.
Miller
B. L.
Freedman
M.
Levine
B.
Episodic memory and regional atrophy in frontotemporal lobar degeneration
Neuropsychologia
 , 
2008
, vol. 
46
 (pg. 
127
-
136
)
Spiers
H. J.
Maguire
E. A.
The neuroscience of remote spatial memory: A tale of two cities
Neuroscience
 , 
2007
, vol. 
149
 (pg. 
7
-
27
)
Spreen
O.
Strauss
E.
A Compendium of Neuropsychological Tests: Administration, norms, and commentary
 , 
1991
New York
Oxford University Press
Strakowski
S. M.
Delbello
M. P.
Adler
C. M.
The functional neuroanatomy of bipolar disorder: A review of neuroimaging findings
Molecular Psychiatry
 , 
2005
, vol. 
10
 (pg. 
105
-
116
)
Strauss
E.
Sherman
E. M. S.
Spreen
O.
A Compendium of Neuropsychological Tests: Administration, norms, and commentary
 , 
2006
USA
Oxford University Press
Tombaugh
T. N.
Trail Making Test A and B: Normative data stratified by age and education
Archives of Clinical Neuropsychology
 , 
2004
, vol. 
19
 (pg. 
203
-
214
)
Vakil
E.
Raz
T.
Levy
D. A.
Probing the brain substrates of cognitive processes responsible for context effects on recognition memory
Aging, Neuropsychology and Cognition
 , 
2010
, vol. 
17
 (pg. 
519
-
544
)
Warrington
E. K.
James
M.
The visual object and space perception battery
 , 
1991
Bury St. Edmunds, UK
Thames Valley Test Company
Wechsler
D.
Wechsler Memory Scale—Third Edition (WMS-III)
 , 
1998
London
The Psychological Corporation
Wedderburn
C.
Wear
H.
Brown
J.
Mason
S. J.
Barker
R. A.
Hodges
J. R.
, et al.  . 
The utility of the Cambridge Behavioural Inventory in neurodegenerative disease
Journal of Neurology, Neurosurgery and Psychiatry
 , 
2008
, vol. 
79
 (pg. 
500
-
503
)