Neuropsychology 2015, Vol. 29, No. 6, 933–939

© 2015 American Psychological Association 0894-4105/15/$12.00 http://dx.doi.org/10.1037/neu0000197

BRIEF REPORT

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Discounting of Future Rewards in Behavioural Variant Frontotemporal Dementia and Alzheimer’s Disease Maxime Bertoux

Leonardo Cruz de Souza

Institut du Cerveau et de la Moelle Epinière & Pitié-Salpêtrière Hospital, Paris, France, and University of Cambridge

Universidade Federal de Minas Gerais

Pablo Zamith

Bruno Dubois

Sciences Po, Paris, France

Institut du Cerveau et de la Moelle Epinière, and PitiéSalpêtrière Hospital, Paris, France

Sacha Bourgeois-Gironde Ecole Normale Supérieure, Paris, France Objective: The clinical differential diagnosis of Alzheimer’s disease (AD) and behavioral-variant frontotemporal dementia (bvFTD) can no longer rely only on episodic memory impairment or executive dysfunctions, as highlighted by recent findings showing that both diseases could present with similar impairments. Objective cognitive tests assessing specific symptoms, such as impulsivity in bvFTD, are thus crucially needed. The aim of this study was to evaluate the differences in impulsivity between bvFTD and AD using a delay-discounting paradigm. Method: An ecological delay-discounting test was administrated to 70 participants including 30 ADs, 20 bvFTD and 20 controls. AD patients were divided according to the severity of the disease into mild or moderate group. The delay-discounting score, reflecting the total percentage of impulsive choice across the entire task, was analyzed for each group. Results: This score showed that bvFTD patients were significantly more impulsive than controls and AD patients at mild or moderate stage. AD patients, regardless of disease stage, did not differ from controls. ROC analyses revealed high and significant area under the curve (AUC, 95% confidence interval) for this score to differentiate bvFTD from AD (0.704) or controls (0.904), or both group (AD ⫹ controls; AUC ⫽ 0.791). Conclusion: The total delay-discounting score provided by our task showed that it could accurately differentiate bvFTD patients from AD and controls. These results support the relevancy of using tests inspired by experimental psychoeconomics and taping into reward processing to increase the distinction between both diseases. Keywords: behavioral variant frontotemporal dementia, Alzheimer’s disease, impulsivity, delaydiscounting, intertemporal choices Supplemental materials: http://dx.doi.org/10.1037/neu0000197.supp

This article was published Online First April 20, 2015. Maxime Bertoux, Institut du Cerveau et de la Moelle Epinière, Paris, France; Institut de la Mémoire et de la Maladie d’Alzheimer, PitiéSalpêtrière Hospital, Paris, France; Department of Clinical Neurosciences, University of Cambridge; Leonardo Cruz de Souza, Universidade Federal de Minas Gerais, Belo-Horizonte; Pablo Zamith, Centre MaxPo, Sciences Po, Paris, France; Bruno Dubois, Institut du Cerveau et de la Moelle Epinière, Institut de la Mémoire et de la Maladie d’Alzheimer, PitiéSalpêtrière Hospital; Sacha Bourgeois-Gironde, Institut Jean Nicod, Ecole Normale Supérieure, Paris, France. We thank Mathias Pessiglione for the design of the task and Philippe Fossati for helpful discussions. We also thank Virginie Czenercki,

Aurélie Funkiewiez, Elodie Guichart-Gomez, Valérie Hahn-Barma, Christina Rogan, and Dalila Samri for performing most of the neuropsychological assessments of the patients included. This work was partially conducted during Maxime Bertoux’s doctoral program, which was supported by the French Ministry of Defense. Maxime Bertoux is currently supported by a Marie Curie-Skłodowska fellowship from the European Commission. Correspondence concerning this article should be addressed to Maxime Bertoux, Department of Clinical Neurosciences Herchel Smith Building, Forvie Site, Addenbrooke’s Hospital, Cambridge, United Kingdom. E-mail: mb2044@ medschl.cam.ac.uk 933

BERTOUX ET AL.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

934

Impulsivity can be defined as a tendency to choose short-term gains over larger but delayed ones (Rachlin, 2000). Although being a natural facet of human behavior (Green, Myerson, & Ostaszewski, 1999), abnormal patterns of impulsivity have been observed in neurodegenerative diseases (Rosen et al., 2005). Together with apathy, lack of empathy, and dietary changes, impulsivity is a prominent symptom of behavioral-variant frontotemporal dementia (bvFTD; Piguet, Hornberger, Mioshi, & Hodges, 2011). Clinical scales evaluating impulsivity have showed good results to distinguish bvFTD from Alzheimer’s disease (AD; Levy, Miller, Cummings, Fairbank, & Craig, 1996). Considering that these two diseases share common symptoms leading to frequent misdiagnosis such as dysexecutive syndrome (Possin et al., 2013), episodic memory impairment (Bertoux et al., 2014; Hornberger, Piguet, Graham, Nestor, & Hodges, 2010), and decision-making difficulties (Bertoux, Funkiewiez, O’Callaghan, Dubois & Hornberger, 2013; Kloeters et al., 2013), objective neuropsychological tests of impulsivity are crucially needed as they could provide relevant information for the differential diagnosis. One of the most common methods to assess impulsivity relies on the delay-discounting paradigm, in which a subject has to choose between an immediate small reward and a larger but delayed one (Mazur, 1987; Madden et al., 1997). Although being widely used, recent studies raised concerns about its lack of ecological validity (Bourgeois-Gironde, 2012; Rick & Loewenstein, 2008). Indeed, classic delay-discounting paradigms involved monetary, hence abstract, payoffs when everyday decisions rely on concrete options. In addition, real-life decisions generally involve a choice between tangible– observed immediate options (e.g., eating a snack, smoking a cigarette, drinking alcohol . . .) and intangible–imagined future situations (having a larger meal, getting addicted, staying healthy . . .). When having real-life intertemporal choices to make, humans thus have to produce a mental representation of the future rewards, whereas the immediate ones are immediately perceived. In addressing these issues, a recent task has been proposed to extend previous intertemporal choice paradigms by showing concrete options with two modes of presentation: some options were accompanied with pictures and thus directly observable, whereas others were only described textually, requiring mental simulation (Lebreton et al., 2013). The first aim of this study is to evaluate the ability of a clinical adaptation of this ecological delay-discounting paradigm to distinguish bvFTD from AD and controls with a single score evaluating the percentage of impulsive choices during the task. Through more detailed analyses, the second aim is to investigate the differences between the ecological experimental condition (when the immediate option is observable and the future option is only written) and the other ones (when both options are observables or written).

Method Participants Seventy participants were recruited at the Alzheimer Institute of the Pitié-Salpêtrière Hospital in Paris (France), resulting in 30 AD patients, 20 bvFTD patients, and 20 healthy controls. Patients fulfilled consensual diagnostic criteria for bvFTD (Rascovsky et al., 2011) or AD (Albert et al., 2011; Dubois et al., 2007). All patients were followed at least 24 month to increase the clinical

confidence of the diagnosis. According to clinical dementia rating (CDR) scores, AD patients were divided into AD-MILD (mild, n ⫽ 15; CDR ⫽ 0.5) and AD-MOD (moderate, n ⫽ 15; CDR ⬎ 0.5) subgroups. Controls had no cognitive complaints and no history of neurologic–psychiatric disorders. All controls had normal cognitive assessment. This study was approved by the local ethics committee. All participants signed informed-consent forms. Data of 55% (n ⫽ 39/70) of the participants were partially presented in a previous study (Lebreton et al., 2013).

Neuropsychological Assessment AD and bvFTD patients underwent a comprehensive neuropsychological assessment (see Table 1), composed from the MiniMental State Examination (MMSE; Folstein, Folstein, & McHugh, 1975), the FAB (Dubois, Slachevsky, Litvan, & Pillon, 2000), the Free and Cued Selective Reminding Test (FCSRT; Grober, Buschke, Crystal, Bang, & Desner, 1988) assessing episodic memory, plus verbal fluency, word denomination, and digit span tests as well as the mini-SEA (Bertoux et al., 2013) assessing social and emotional cognition.

Ecological Intertemporal Choices Task The task was programmed on Mac OS X using Matlab 11 (MathWorks, Cambridge, England; (see Figure 1). Participants were seated in front of a computer screen and had to choose between two options: a less pleasant but immediate reward versus a more pleasant but delayed (1 month, 1 year, or 10 years) one. Both options were presented at the same time on the screen, one in the left side, the other in the right side, and participants had to state verbally which reward they chose. Instructions were repeated for each choice. The task was composed from two sets of stimuli: Observed (Obs) options, which were displayed with pictures and a short sentence, and simulated (Sim) options, which were described with words only. Participants underwent two sessions composed of 72 choice trials divided into three conditions: Obs/Sim (n ⫽ 36), Sim/Sim (n ⫽ 18), and Obs/Obs (n ⫽ 18). The first condition corresponds to everyday life situations where an immediate reward is observed, whereas the delayed one requires mental simulation (ecological condition). Each option belonged to different domains (food, culture, or sport). The two factors (delay and domain) were crossed to yield different combinations, which were presented in a pseudorandomized order and independently of the side of the screen. The total number of choices was fixed (n ⫽ 144). Data extracted from this task was the percentage of nonimpulsive choices (i.e., the number of immediate choices divided by the total number, multiplied by 100) for each group for (a) the whole task (delay-discounting score) and (b) each delays, conditions, and domains. A more detailed description of the task is presented in the Appendix.

Statistical Analyses The data were analyzed using SPSS 20 (SPSS Inc., Chicago, Illinois) and MedCalc (http://www.medcalc.org) software. Before any analysis, variables were plotted and checked for normality of distribution using the Shapiro-Wilk test. All variables followed

DISCOUNTING OF REWARD IN AD & BVFTD

935

Table 1 Demographics, Neuropsychological Data, and Percentage of Nonimpulsive Choice at the Ecological Intertemporal Choices Task for Each Group According to the delays

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Demographics Age Education Sex (W/M) Neuropsychological examination MMSE (/30) FAB (/18) FCSRT Free (/48) FCSRT total (/48) Sem. verbal fluency Morp. verbal fluency Mini-SEA (/30) Praxis (%) Word denomination Digit forward Digit backward Ecological intertemporal choices task (% of nonimpulsive choice) Immediate reward versus 1 month 1 year 10 years M

Controls

AD-MILD

AD-MOD

bvFTD

73.05 (7.58)a 9.1 (3.80) 10/10

70.93 (5.47)a 9.6 (4.15) 7/8

77.50 (7.24)a 9.6 (4.61) 5/10

65.6 (7.57)b,c 10.1 (3.87) 9/11

28.95 (0.83)a,b,c 17.50 (0.51)a,b,c NA NA NA NA NA NA NA NA NA

26.47 (1.30)a,c 15.73 (1.79)a 7.8 (7.16) 26.7 (5.69) 15.55 (3.36) 12.00 (3.07)a 25.97 (2.01)a 86 (21.42) 95.51 (4.91) 5.93 (1.21) 4.43 (0.76)a

20.5 (4.20)b 14.07 (1.69) 4.75 (5.10)a 12.25 (13.60)a 16.44 (7.27) 12.67 (6.56) 25.56 (1.35)ⴱ 81 (23.27) 85.94 (21.29) 5.50 (1.57) 4.17 (1.47)

23.65 (2.60)b 12.65 (2.08)b 15.13 (6.92)c 33.67 (13.14)c 10.33 (4.69) 6.00 (4.07)b 14.30 (3.71)b,c 97 (4.28) 94.22 (7.10) 5.00 (0.82) 3.25 (1.24)

69.58 (11.35)a,b,c 68.54 (18.81)a,b,c 35.54 (29.03) 57.94 (14.44)a

52.87 (15.93) 46.67 (15.63)a 43.24 (21.98)a 47.59 (15.63)a

55.74 (12.94) 52.87 (9.17)a 35.19 (16.58) 47.93 (9.87)a

44.59 (21.20) 35.09 (17.06)b,c 21.93 (16.57)b 33.86 (15.54)b,c

Note. AD ⫽ Alzheimer’s dementia; bvFTD ⫽ behavioral-variant frontotemporal dementia; M (SD). MMSE ⫽ Mini-Mental State Examination; FAB ⫽ Frontal Assessment Battery; FCSRT ⫽ Free and Cued Selective Reminding Test; Mini-SEA ⫽ short version of The Social Cognition and Emotional Assessment. Significant differences: a Vs bvFTD. b Vs AD-MILD. c Vs AD-MOD.

nonnormal distribution. Demographic and neuropsychological data were thus analyzed with nonparametric tests (Mann–Whitney U). The Spearman test was used to study the correlations with these data. To conduct multifactorial analysis of variance (ANOVA) on the nonnormally distributed experimental data (scores from the delay-discounting test), an arcsine transformation was performed on these data. The arcsine transformation is a procedure somewhat similar to logarithmic transformation used in previous studies in the field (e.g., Hornberger, Piguet, Graham, Nestor, & Hodges, 2010) and allowed us to consider and analyze values that are equal to zero. Such values in our study are not because of the lack of sensitivity of the outcome measures and are not cases of missed data. We also believe that the arcsine transformation is the most relevant choice as our experimental data are proportions or percentages. A first analysis was conducted on the total delay-discounting score, aiming to investigate the differences between the participant groups. To test the accuracy of this score to classify each participant in the appropriate group, we also conducted binary logistic regressions with Enter method. Then, received operating characteristics (ROC) curve analyses were performed to evaluate the diagnosis ability of this score. The area under the curve (AUC) was used as a measure of the overall performance of each ROC curve. A second analysis was focused on each factor of the intertemporal choices task to investigate the qualitative profile of each group and the influence of each factor on the participant’s performance. A 4 ⫻ 3 ⫻ 3 ⫻ 3 ANOVA was used to examine the main effects of group, delay, domain, and condition variables and their interactions, following by Bonferonni post hoc test, which serve as

a correction for multiple comparisons, to assess the significance of the effects. Student t test for unique sample was then used to test the differences across within-subject factors (delays, domains, conditions), and Student t test for independent samples was used to assess differences across between-subjects factor (groups). Cohen’s d was retained to estimate effect sizes.

Results All results were corrected for multiple comparisons using the Bonferonni correction and are reported with a 95% confidence interval.

Demographics and Neuropsychological Data bvFTD patients were younger than controls (Z ⫽ 2.67; p ⬍ 10⫺2), AD-MILD (Z ⫽ 2.35; p ⫽ .017), and AD-MOD (Z ⫽ 3.47; p ⬍ 10⫺4) groups. The four groups did not differ on sex or level of education (see Table 1). AD-MILD patients have higher MMSE scores than bvFTD (Z ⫽ 3.47; p ⬍ 10⫺3) and AD-MOD (Z ⫽ 4.23; p ⬍ 10⫺5) patients, as well as a higher morphological fluency (Z ⫽ 3.00; p ⬍ 10⫺3), FAB (Z ⫽ 3.66; p ⬍ 10⫺4) and backward digit span (Z ⫽ 3.01; p ⬍ 10⫺3) scores compared with bvFTD. bvFTD patients have higher free recall (Z ⫽ 3.36; p ⬍ 10⫺3) and total recall (Z ⫽ 3.33; p ⬍ 10⫺3) memory scores than AD-MOD patients. They have lower mini-SEA scores compared with both AD-MILD (Z ⫽ 2.68; p ⬍ 10⫺3) and AD-MOD patients (Z ⫽ 3.57; p ⬍ 10⫺4).

BERTOUX ET AL.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

936

Figure 1. A: Sample of choices from the ecological intertemporal choices task with variation of delay, domain, and condition. On the left, rewards from food domain were proposed in the observed/simulated (Obs/Sim) condition with 1 month of interval. In the middle, rewards from culture domain were proposed in the Sim/Sim condition with 1 year of interval. On the right, rewards from sport domain were proposed in the Obs/Obs condition with 10 years of interval. B: Delay-discounting curve, with percentage of nonimpulsive choice in ordinate and delay in abscissa. C: Mean percentage of nonimpulsive choice (delay discounting score) on the entire task for each groups. bvFTD ⫽ behavioral variant frontotemporal dementia; AD ⫽ Alzheimer’s disease; MCI ⫽ mild AD; MOD ⫽ moderate. See the online article for the color version of this figure.

Ecological Intertemporal Choices Task Delay-discounting score (total percentage of impulsive choices). Differences between groups showed the following pattern: bvFTD ⬎ AD-MILD ⫽ AD-MOD ⫽ Controls (see Table 1, Figure 1). bvFTD patients were significantly more impulsive than mild AD (t ⫽ 2.64, p ⬍ .01, d ⫽ 0.86), moderate AD (t ⫽ 3.23, p ⬍ 10⫺3, d ⫽ 1.08), and controls (t ⫽ 5.06, p ⬍ 10⫺5, d ⫽ 1.56). AD subgroups did not differ from controls. Binary logistic regression models show that the delay-discounting score could adequately distinguish bvFTD from AD (regardless of disease severity) and controls with respectively 75.5% and 82.1% of accuracy. ROC analyses revealed high and significant area under the curve (AUC, 95% confidence interval) for this value differentiating bvFTD from AD (0.704) or controls (0.904), or both group (AD ⫹ Controls; AUC ⫽ 0.791). The same results were observed when age was considered as a nuisance covariate. Qualitative profiles (detailed results). ANOVA showed significant effects of group (D ⫽ 68.71, p ⬍ 10⫺7), delay (D ⫽ 91.78; p ⬍ 10⫺7), domain (D ⫽ 23.37, p ⬍ 10⫺7) and condition (D ⫽ 7.99, p ⬍ 10⫺3) as well as significant interaction effects: Group ⫻ Delay (D ⫽ 8.69, p ⬍ 10⫺7), Group ⫻ Domain (D ⫽ 4.04, p ⬍ 103), and Condition ⫻ Domain (D ⫽ 6.28, p ⬍ .10⫺4). Group effect. This result was detailed in the Delaydiscounting score section.

Delay effect. bvFTD patients were more impulsive for 10years (t ⫽ 3.75; p ⬍ 103) than for 1 year and for 1 year than for 1 month (t ⫽ 3.42; p ⬍ 103). For AD-MOD and controls, significant differences between 10 years and 1 year were observed (t ⫽ 4.79, p ⬍ 104 and t ⫽ 4.69, p ⬍ 104, respectively) but not between 1 year and 1 month. AD-MILD showed differences between 1 month and 1 year (t ⫽ 2.58, p ⫽ .02) and between 1 month and 10 years (t ⫽ 2.06, p ⫽ .05), but these differences did not pass correction for multiple comparisons. Domain effect. Results for each domain are presented in the Appendix (see Table A2). Participants were generally more significantly impulsive for culture than for sport (t ⫽ 5.26, p ⬍ 106) and for sport than for food (t ⫽ 3.56, p ⬍ 103). There was no difference between groups. Condition effect. Results for each condition are presented in the Appendix (see Table A3). There was no difference between “Obs/Sim” and “Sim/Sim,” but participants were significantly less impulsive (t ⫽ 2.63, p ⫽ 10⫺2) in the Obs/Obs condition. Given that the Obs/Sim condition was considered to be ecological compared with the others, further analyses were conducted to investigate the differences between the three conditions within each group and across groups. No significant difference was observed across the conditions within each group and between the four groups. When AD-MILD and AD-MOD were considered together,

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

DISCOUNTING OF REWARD IN AD & BVFTD

the same result was observed. In this last case, binary logistic regression only showed slight differences between the conditions to discriminate bvFTD and AD patients (Obs/Sim ⫽ 73.5%; Obs/ Obs ⫽ 69.4%; Sim/Sim ⫽ 69.4%), and AUC values were not significantly different (0.250, 0.269, 0.337, respectively). Group ⴛ Delay interaction. For “now versus 1 month,” bvFTD (t ⫽ 4.97, p ⬍ 10⫺4, d ⫽ 1.47), AD-MILD (t ⫽ 3.67, p ⬍ 10⫺3, d ⫽ 1.21), and AD-MOD patients (t ⫽ 3.75, p ⬍ 10⫺3, d ⫽ 1.14) were more impulsive than controls with no significant differences between patients’ groups. For “now versus 1 year,” bvFTD were significantly more impulsive than controls (t ⫽ 5.72, p ⬍ 10⫺5, d ⫽ 1.86) as well as moderate AD (t ⫽ 3.76, p ⬍ 10⫺3, d ⫽ 1.51). Both mild AD (t ⫽ 3.62, p ⬍ 10⫺3, d ⫽ 1.09) and moderate AD were also significantly more impulsive than controls (t ⫽ 3.69, p ⬍ 10⫺3, d ⫽ 1.48). There was no difference between AD-MILD and AD-MOD. For “now versus 10 years,” there was no significant difference between bvFTD, AD-MOD, and controls, but AD-MILD was less impulsive (t ⫽ 3.14, p ⬍ 10⫺3, d ⫽ 1.09) than bvFTD. Group ⴛ Domain interaction. bvFTD were significantly more impulsive than AD-MOD in sport (t ⫽ 3.25, p ⬍ 10⫺3) and culture (t ⫽ 2.99, p ⬍ 10⫺3) rewards. They were also more impulsive than AD-MILD for sport (t ⫽ 5.18, p ⬍ 10⫺5) and controls in all domains (all ts ⬎ 4.4, p ⬍ 10⫺4). There was no difference in any domain in both AD subgroups. There were no differences between controls and AD-MILD or AD-MOD in any domains. Condition ⴛ Domain interaction. The main effect of condition was only observed in the food domain, in which reward in the Obs/Obs condition were less devaluated than in the Obs/Sim (t ⫽ 5.97, p ⬍ 10⫺7) and Sim/Sim (t ⫽ 4.29, p ⬍ 10⫺5) conditions. Further analyses revealed that this pattern was only true when all participants were considered together. Intragroup correlations with neuropsychological variables. The percentage of nonimpulsive choices for the delay-discounting task was not correlated with any of the cognitive data or demographics, as shown in the Appendix (see Tables A4, A5, and A6).

Discussion This study showed that an ecological delay-discounting paradigm could accurately differentiate bvFTD patients from AD and controls. To the best of our knowledge, this is the first time that impulsivity in bvFTD is assessed by an objective test based on intertemporal decisions. The delay-discounting score provided by our task showed that bvFTD patients were more likely to choose less pleasant but immediate rewards versus more pleasant but delayed rewards comparatively with AD patients or controls. In other words, bvFTD patients were significantly more impulsive than other participants. Qualitative analyses revealed that bvFTD have a unique choice profile. Although controls and AD patients only discounted the big rewards associated with the longest delay (10 years), bvFTD patients consistently preferred the small rewards, whenever the big ones were associated with a 1-month, 1-year, or 10-years delay. When splitting the AD group according to the severity of the disease, we observed different profiles for mild and moderate patients. More precisely, although an effect of the delay on impulsivity was observed in AD-MOD patients, AD-MILD showed

937

no significant effect of delay on their choice profiles although statistical trends were observed. As shown by the values and the discounting curve from the statistical results and Figure 1, we believe that we failed to capture the effect of delay in this group because of its size. Another interesting result is the absence of difference in impulsivity between controls, AD-MOD and bvFTD when an immediate reward was faced to a 10-year delayed bigger one. The discount of future reward is a natural and normal process (Green et al., 1999), depending on objective (monetary value of the reward) and subjective variables (tastes, foresight, strength of will, habit, uncertainty . . .). One possible explanation for this absence of difference is that a 70-year-old person may consider that he or she won’t be in a sufficient physical condition to get the reward in 10 years, considering that some of these rewards involved to travel in a foreign country (i.e., to visit the Great Wall of China.). Another likely explanation is that, in this task, the size of the biggest rewards may have been insufficiently high to not be discounted as dramatically by the controls, considering that 10 years is very large delay, usually associated with larger reward in past studies (Holt et al., 2012). It should be noted that the delays of the task were selected after analyzing the data of a pilot study performed on controls (n ⫽ 12). We chose these delays at the conclusion of this study because (a) participants’ impulsivity increased with the rise of the delay and (b) for each delay, participants showed a mixed profile of impulsive and nonimpulsive choices, therefore avoiding a floor or ceiling effect. In any case, further studies on older participants should take into account this absence of difference on the 10-year delay. The second aim of this study was to explore the differences between the ecological condition (when a participant has to choose between a tangible and immediate reward and an imagined future reward) and the two others (when both rewards were observed or imagined). It appeared that there was no difference between the groups across the three conditions. Although these results are not supporting the fact that an ecological condition is more relevant to assess impulsivity in a clinical setting, we still believe that intangibility in intertemporal choices is a crucial issue (Rick & Loewenstein, 2008) and that future studies have to address this question. The absence of correlation between delay-discounting and cognitive scores in this study showed that this kind of paradigm assesses cognitive processes that are not evaluated by memory, executive, language or social-cognition neuropsychological tests. Despite its promising findings, our study did not cross-correlate behavioral findings with neuroimaging data such as voxel-based morphometry, thus, we cannot know for sure which brain areas contributed specifically to these findings. However, a previous fMRI study with this task has shown the involvement of vmPFC and more dorsolateral regions of the prefrontal cortex during valuations of the presented options and the activation of dorsomedian and lateral prefrontal regions during the choices, as well as the left hippocampus (Lebreton et al., 2013). More generally, in past studies, delay-discounting paradigms have been shown to involve mainly the VMPFC and striatum during the immediate choices as well as the dorsolateral PFC during delayed choices (McClure, Laibson, Loewenstein, & Cohen, 2004), which are considered as key nodes in the human brain valuation system (Kable & Glimcher, 2007; Rolls, 2006). Although the hippocampus atrophy is commonly observed in AD as well as in FTD (de Souza et al.,

BERTOUX ET AL.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

938

2013), both the vmPFC and the striatum have been showed to be atrophic specifically in bvFTD and not in AD, and their dysfunctions have been consistently linked to a lack of inhibition in bvFTD (Hornberger, Geng, & Hodges, 2011; O’Callaghan, Bertoux, & Hornberger, 2014). Whether these impulsivity findings in bvFTD are because of a deficit in inhibition per se or to the difficulty to imagine future situations remains an open question. Indeed, recent studies have shown that episodic future thinking is impaired in bvFTD (Irish et al., 2013). An intertemporal choice could require a mental time travel to simulate the delayed option (Boyer, 2008), and, along these lines, the hypothesis of a disruption of the episodic future-thinking network would be coherent with the present findings. Unfortunately, data is lacking to answer this question, as the certainty and quality of the mental projection itself was not assessed. Some of the reported findings may be nonintuitive. For example, the absence of impulsivity differences in the “food” domain between bvFTD and other groups may appear surprising considering the hyperorality or dietary changes observed in bvFTD. Also, we don’t explain that only mild AD patients, and not controls, were less impulsive than bvFTD for the longest delay. We believed that these effects may be because of the limited size of our sample. These results need to be replicated in other independent and larger series. Another factor that may limit the final interpretation of the present data is that patients did not have an autopsy-confirmed diagnosis. However, consensual criteria were used to establish the clinical diagnosis, and all patients were followed for at least 24 months in a center with high expertise in the field of cognitive neurology, thus increasing the confidence of the clinical diagnosis. Development of clinical delay-discounting tasks could undeniably have an important clinical interest in the objective evaluation of impulsivity, a major symptom of bvFTD, which is, to date, mainly based on clinical scale dependant on career. For the first time, using a single score based on the number of immediate reward chosen during an intertemporal task, we showed that bvFTD were more impulsive than AD and controls. Intertemporal decision tasks can therefore be used to assess impulsivity and to detect dysfunctions in the reward system network in neurodegenerative diseases, which in turn can provide markers for clinical and behavioral interventions. Finally, the results of our study support the idea that the objective assessment of functions related to the vmPFC, such as social cognition or behavioral processes, could further increase the sensitivity and specificity of bvFTD diagnosis and should be taken into account in the next revision of clinical diagnostic criteria.

References Albert, M. S., DeKosky, S. T., Dickson, D., Dubois, B., Feldman, H. H., Fox, N. C., . . . Phelps, C. H. (2011). The diagnosis of mild cognitive impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer’s & Dementia, 7, 270 –279. http://dx.doi.org/10.1016/j.jalz.2011.03.008 Bertoux, M., de Souza, L. C., Corlier, F., Lamari, F., Bottlaender, M., Dubois, B., & Sarazin, M. (2014). Two distinct amnesic profiles in behavioral variant frontotemporal dementia. Biological Psychiatry, 75, 582–588. http://dx.doi.org/10.1016/j.biopsych.2013.08.017 Bertoux, M., Funkiewiez, A., O’Callaghan, C., Dubois, B., & Hornberger, M. (2013). Sensitivity and specificity of ventromedial prefrontal cortex

tests in behavioral variant frontotemporal dementia. Alzheimer’s & Dementia, 9(Suppl.), S84 –S94. http://dx.doi.org/10.1016/j.jalz.2012.09 .010 Bourgeois-Gironde, S. (2012). Optimal short-sighted rules. Frontiers in Neuroscience, 6, 129. http://dx.doi.org/10.3389/fnins.2012.00129 Boyer, P. (2008). Evolutionary economics of mental time travel? Trends in Cognitive Sciences, 12, 219 –224. http://dx.doi.org/10.1016/j.tics.2008 .03.003 de Souza, L. C., Chupin, M., Bertoux, M., Lehéricy, S., Dubois, B., Lamari, F., . . . Sarazin, M. (2013). Is hippocampal volume a good marker to differentiate Alzheimer’s disease from frontotemporal dementia? Journal of Alzheimer’s Disease, 36, 57– 66. Dubois, B., Feldman, H. H., Jacova, C., Dekosky, S. T., Barberger-Gateau, P., Cummings, J., . . . Scheltens, P. (2007). Research criteria for the diagnosis of Alzheimer’s disease: Revising the NINCDS-ADRDA criteria. The Lancet. Neurology, 6, 734 –746. Dubois, B., Slachevsky, A., Litvan, I., & Pillon, B. (2000). The FAB: A Frontal Assessment Battery at bedside. Neurology, 55, 1621–1626. http://dx.doi.org/10.1212/WNL.55.11.1621 Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). “Mini-mental state.” A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189 –198. http://dx .doi.org/10.1016/0022-3956(75)90026-6 Green, L., Myerson, J., & Ostaszewski, P. (1999). Discounting of delayed rewards across the life span: Age differences in individual discounting functions. Behavioural Processes, 46, 89 –96. http://dx.doi.org/10.1016/ S0376-6357(99)00021-2 Grober, E., Buschke, H., Crystal, H., Bang, S., & Dresner, R. (1988). Screening for dementia by memory testing. Neurology, 38, 900 –903. http://dx.doi.org/10.1212/WNL.38.6.900 Holt, D. D., Green, L., & Myerson, J. (2012). Estimating the subjective value of future rewards: Comparison of adjusting-amount and adjustingdelay procedures. Behavioural Processes, 90, 302–310. http://dx.doi .org/10.1016/j.beproc.2012.03.003 Hornberger, M., Geng, J., & Hodges, J. R. (2011). Convergent grey and white matter evidence of orbitofrontal cortex changes related to disinhibition in behavioural variant frontotemporal dementia. Brain: A Journal of Neurology, 134, 2502–2512. http://dx.doi.org/10.1093/brain/ awr173 Hornberger, M., Piguet, O., Graham, A. J., Nestor, P. J., & Hodges, J. R. (2010). How preserved is episodic memory in behavioral variant frontotemporal dementia? Neurology, 74, 472– 479. http://dx.doi.org/ 10.1212/WNL.0b013e3181cef85d Irish, M., Hodges, J. R., & Piguet, O. (2013). Episodic future thinking is impaired in the behavioural variant of frontotemporal dementia. Cortex, 49, 2377–2388. http://dx.doi.org/10.1016/j.cortex.2013.03.002 Kable, J. W., & Glimcher, P. W. (2007). The neural correlates of subjective value during intertemporal choice. Nature Neuroscience, 10, 1625– 1633. http://dx.doi.org/10.1038/nn2007 Kloeters, S., Bertoux, M., O’Callaghan, C., Hodges, J. R., & Hornberger, M. (2013). Money for nothing—Atrophy correlates of gambling decision making in behavioural variant frontotemporal dementia and Alzheimer’s disease. Neuroimage: Clinical, 2, 263–272. http://dx.doi.org/ 10.1016/j.nicl.2013.01.011 Lebreton, M., Bertoux, M., Boutet, C., Lehericy, S., Dubois, B., Fossati, P., & Pessiglione, M. (2013). A critical role for the hippocampus in the valuation of imagined outcomes. PLoS Biology, 11(10), e1001684. http://dx.doi.org/10.1371/journal.pbio.1001684 Levy, M. L., Miller, B. L., Cummings, J. L., Fairbanks, L. A., & Craig, A. (1996). Alzheimer disease and frontotemporal dementias. Behavioral distinctions. Archives of Neurology, 53, 687– 690. http://dx.doi.org/ 10.1001/archneur.1996.00550070129021 Madden, G. J., Petry, N. M., Badger, G. J., & Bickel, W. K. (1997). Impulsive and self-control choices in opioid-dependent patients and

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

DISCOUNTING OF REWARD IN AD & BVFTD non-drug-using control patients: Drug and monetary rewards. Experimental and Clinical Psychopharmacology, 5, 256 –262. http://dx.doi .org/10.1037/1064-1297.5.3.256 Mazur, J. E. (1987). An adjusting procedure for studying delayed reinforcement. In M. L. Commons, J. E. Mazur, J. A. Nevin, & H. Rachlin (Eds.), Quantitative analyses of behavior: The effect of delay and of intervening events on reinforcement value (Vol. 5, pp. 55–73). Hillsdale, NJ: Erlbaum. McClure, S. M., Laibson, D. I., Loewenstein, G., & Cohen, J. D. (2004, October 15). Separate neural systems value immediate and delayed monetary rewards. Science, 306, 503–507. http://dx.doi.org/10.1126/ science.1100907 O’Callaghan, C., Bertoux, M., & Hornberger, M. (2014). Beyond and below the cortex: The contribution of striatal dysfunction to cognition and behaviour in neurodegeneration. Journal of Neurology, Neurosurgery & Psychiatry, 85, 371–378. http://dx.doi.org/10.1136/jnnp-2012304558 Piguet, O., Hornberger, M., Mioshi, E., & Hodges, J. R. (2011). Behavioural-variant frontotemporal dementia: Diagnosis, clinical staging, and management. The Lancet. Neurology, 10, 162–172. http://dx .doi.org/10.1016/S1474-4422(10)70299-4 Possin, K. L., Feigenbaum, D., Rankin, K. P., Smith, G. E., Boxer, A. L., Wood, K., . . . Kramer, J. H. (2013). Dissociable executive functions in

939

behavioral variant frontotemporal and Alzheimer dementias. Neurology, 80, 2180 –2185. http://dx.doi.org/10.1212/WNL.0b013e318296e940 Rachlin, H. (2000). The science of self-control. Cambridge, MA: Harvard University Press. Rascovsky, K., Hodges, J. R., Knopman, D., Mendez, M. F., Kramer, J. H., Neuhaus, J., . . . Miller, B. L. (2011). Sensitivity of revised diagnostic criteria for the behavioural variant of frontotemporal dementia. Brain: A Journal of Neurology, 134, 2456 –2477. Rick, S., & Loewenstein, G. (2008). Intangibility in intertemporal choice. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 363, 3814 –3824. http://dx.doi.org/10.1098/rstb .2008.0150 Rolls, E. T. (2006). The neurophysiology and functions of the orbitofrontal cortex. In D. H. Zald & S. Rauch (Eds.), The orbitofrontal cortex (pp. 95–124). New York, NY: Oxford University Press. http://dx.doi.org/ 10.1093/acprof:oso/9780198565741.003.0005 Rosen, H. J., Allison, S. C., Schauer, G. F., Gorno-Tempini, M. L., Weiner, M. W., & Miller, B. L. (2005). Neuroanatomical correlates of behavioural disorders in dementia. Brain: A Journal of Neurology, 128(Pt. 11), 2612–2625. http://dx.doi.org/10.1093/brain/awh628

Received November 4, 2014 Revision received February 9, 2015 Accepted February 9, 2015 䡲

Discounting of future rewards in behavioural variant frontotemporal dementia and Alzheimer's disease.

The clinical differential diagnosis of Alzheimer's disease (AD) and behavioral-variant frontotemporal dementia (bvFTD) can no longer rely only on epis...
768KB Sizes 1 Downloads 8 Views