(PDF) Allocentric coordinate spatial representations are impaired in aMCI and Alzheimer’s disease patients

Journal Pre-proof Allocentric coordinate spatial representations are impaired in aMCI and Alzheimer’s Disease Patients Gennaro Ruggiero (Conceptualization) (Formal analysis) (Investigation) (Methodology)<ce:contributor-role>Writing - original draft) (Writing - review and editing), Francesco Ruotolo (Data curation) (Investigation) (Formal analysis) (Writing - review and editing), Alessandro Iavarone (Resources) (Methodology) (Data curation) (Writing - review and editing), Tina Iachini (Data curation) (Supervision) (Writing - review and editing) PII: S0166-4328(20)30492-7 DOI: https://doi.org/10.1016/j.bbr.2020.112793 Reference: BBR 112793 To appear in: Behavioural Brain Research Received Date: 24 February 2020 Revised Date: 19 June 2020 Accepted Date: 27 June 2020 Please cite this article as: Ruggiero G, Ruotolo F, Iavarone A, Iachini T, Allocentric coordinate spatial representations are impaired in aMCI and Alzheimer’s Disease Patients, Behavioural Brain Research (2020), doi: https://doi.org/10.1016/j.bbr.2020.112793 This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 Published by Elsevier. Allocentric coordinate spatial representations are impaired in aMCI and Alzheimer’s Disease Patients Gennaro Ruggiero*a, Francesco Ruotoloa, Alessandro Iavaroneb, and Tina Iachinia of ro a Laboratory of Cognitive Science and Immersive Virtual Reality, CS-IVR, Department of Psychology, University of Campania L. Vanvitelli, Caserta, Italy b Naples, Italy -p Laboratory of Clinical Neuropsychology, Neurological Unit of “Ospedali dei Colli”, re lP na Corresponding author: ur Gennaro Ruggiero, Ph.D. Laboratory of Cognitive Science and Immersive Virtual Reality, CS-IVR Jo Department of Psychology, University of Campania L. Vanvitelli Viale Ellittico, 31, 81100, Caserta, Italy Tel +39-0823-275329 Fax +39-0823-274759 E-mail:

Highlights 1 -We test ego-/allo-centric and categorical/coordinate encodings in aMCI and AD patients -How is the allocentric deficit of aMCI/AD linked to categorical/coordinate relations? -Subjects give right/left and distance based judgments combined with reference frames -aMCI and AD groups show a selective allocentric-coordinate not categorical deficit -Conversion from normal aging to AD may be traced in allocentric distance-based deficits of Abstract ro Research has reported deficits in egocentric (subject-to-object) and mainly allocentric (object- to-object) spatial representations in the early stages of the Alzheimer’s disease (eAD). To -p identify early cognitive signs of neurodegenerative conversion, several studies have shown alterations in both reference frames, especially the allocentric ones in amnestic-Mild re Cognitive Impairment (aMCI) and eAD patients. However, egocentric and allocentric spatial frames of reference are intrinsically connected with coordinate (metric/variant) and lP categorical (non-metric/invariant) spatial relations. This raises the question of whether allocentric deficit found to detect the conversion from aMCI to dementia is differently affected when combined with categorical or coordinate spatial relations. Here, we compared na eAD and aMCI patients to Normal Controls (NC) on the Ego-Allo/Cat-Coor spatial memory task. Participants memorized triad of objects and then were asked to provide right/left (i.e. categorical) and distance based (i.e. coordinate) judgments according to an egocentric or ur allocentric reference frame. Results showed a selective deficit of coordinate, but not categorical, allocentric judgments in both aMCI and eAD patients as compared to NC group. Jo These results suggest that a sign of the departure from normal/healthy aging towards the AD may be traced in elderly people’s inability to represent and compared distances among elements in the space. 2 Keywords: egocentric-allocentric representations; aMCI; eAD; categorical-coordinate spatial relations; visuo-spatial memory disorders. of ro INTRODUCTION The ability to mentally represent places and to efficiently navigate through them requires the -p integrity of both allocentric (object-to-object, e.g. the school building is 100 meters from the bank) and egocentric (subject-to-object, e.g. the bank is on my right) frames of reference [1-3]. Spatial re disorders such as topographical disorientation characterized by getting lost or wandering, have often been described in the early stages of Alzheimer’s disease (AD) [1; 4-7]. In the attempt to identify lP prodromal cognitive signs of AD onset, much research has focused on amnestic Mild Cognitive Impairment (aMCI), a pre-clinical condition characterized by difficulties in various cognitive non- memory domains and at high risk of neurodegenerative progression [8]. It has been suggested that na the decline of the visual-spatial abilities in aMCI may represent an early marker of the AD conversion. In fact, visuospatial deficits may even precede by ten years episodic and semantic memory changes usually observed in AD [9-10]. ur Several studies have reported that AD patients may exhibit deficits in both reference frames, with special regard to the allocentric ones [5; 11-16]. Egocentric disorders are associated with Jo impairments in associative and parietal areas [17], whereas allocentric impairments are associated with early hippocampal atrophy. Since the hippocampal atrophy has been also observed in aMCI patients, it has been proposed that allocentric deficits may represent an early marker of the neurodegenerative conversion from aMCI to AD [6; 17–26]. In the search for earlier cognitive markers of AD conversion, there are still unexplored issues concerning visuo-spatial memory and specifically the allocentric deficits. Research has indeed observed that the distinction between egocentric and allocentric spatial representations is closely 3 linked to the distinction between coordinate and categorical spatial relations [27]. Coordinate spatial relations are based on a fine-grained metric code that allows for precise distance discrimination between different positions, such as the object is closer to me or to the window. Instead, categorical spatial relations are based on a more abstract nonmetric code, such as right/left, above/below) [27- 30]. We cannot process metric or nonmetric spatial relations without specifying a frame of reference and vice versa. Examples of this combination can be drawn from our daily spatial behaviors that are commonly based on spatial relations such as “the street closer to me (egocentric- coordinate) and/or on my right (egocentric-categorical)” or “the building closer to the bridge (allocentric-coordinate) and/or on the left of the bridge (allocentric-categorical)”. Therefore, the egocentric/allocentric and the categorical/coordinate components seem likely to reflect a flexible, of complex, and interactive organization that it modulated by the spatial task at hand [e.g. 28-31]. Importantly, these spatial representations seem to be supported by differently lateralized neural networks. Categorical and coordinate egocentric representations are supported from a fronto- ro parietal network, more on the right side for the coordinate judgments. Instead, categorical and coordinate allocentric representations are supported by bilateral occipito-temporal areas, with an -p involvement of some right-sided parietal areas for the coordinate spatial relations [32]. At behavioural level, coordinate judgments are more difficult as compared to the categorical ones, re especially when combined with an allocentric reference frame [e.g. 33-36]. Compared to the huge amount of studies that have investigated the effect of healthy and lP pathological aging on the processing of egocentric and allocentric information, very few studies have investigated the role of categorical and coordinated spatial relationships and none the combination between frames of reference and spatial relations during ageing. As for categorical and na coordinated spatial relation, for example, Bruyer and colleagues [36] reported that aging has a detrimental effect on coordinate spatial relations processing. In line with this, Antonova and colleagues [37] have shown that young people have an increased activity in the right hemisphere, ur more involved in coordinate judgments, whereas older adults ranging between 64 and 79 years show stronger activation in the left hemisphere, more involved in categorical judgments [38]. Jo Finally, and more importantly, Hampstead and colleagues [39] compared brain activity of aMCI patients and healthy elderly individuals during an objects location memory task. Results showed a reduced activity in right frontal and temporal brain areas of aMCI patients and this was related to their poor performance at the task. According to the authors, the higher right brain activity had facilitated the use of coordinate spatial processing in healthy elderly people, but this right- hemispheric function gradually diminished as an individual progressed to aMCI. 4 Capitalizing on the above mentioned studies, it is possible to hypothesize that a difficulty in processing coordinate spatial information according to an allocentric frame of reference may represent an early marker of the conversion from aMCI to AD. After a general neuropsychological assessment, heathy elderly participants (normal control group: NC), aMCI and AD participants were asked to perform the Ego-Allo/Cat-Coor spatial memory task. This task is based on spatial localization judgments that combine egocentric and allocentric frames of reference with categorical and coordinate spatial relations. For instance, it requires the encoding of distances (coordinate) or relations (categorical) with respect to the participant’s body (egocentric: the object is close to me or on my right) or to an external object (allocentric: the object is close to the cube or on its left). This experimental paradigm has already of been used to assess spatial memory in healthy adults [34-35; 40], brain damaged patients [41], blind people [42-44], in a fMRI study [32; 45], and has proved its efficacy in inducing a specific involvement of spatial frames of reference. ro In the present study, participants memorized triads of three-dimensional (3D) geometrical objects and, after a delay, they provided the combined egocentric-allocentric and coordinate-categorical -p spatial judgments. We expect that aMCI patients, similarly to eAD patients, should reveal a specific difficulty with the allocentric coordinate encoding. re MATERIALS AND METHOD lP Participants Thirty-eight participants took part in the study on a voluntary basis: 10 aMCI (3 women; range of na age= 60-85 years), 8 eAD (5 women, range of age= 60-85 years), 20 NC (9 women, range of age=61-86 years). They all were right-handed and had normal or corrected to normal sight. eAD and aMCI patients were recruited from a panel of patients of the Laboratory of Clinical ur Neuropsychology, Neurological Unit of “Ospedali dei Colli” (Naples, Italy). Exclusion criteria for all groups were the following: severe presence of head trauma or prior neurological disorders, major Jo medical diseases, alcohol or drug abuse and psychiatric disorders. All participants gave their informed consent to take part in the study. They were unaware of the rationale of the study. The study was conform to the local Ethics Committee requirements (v. CdD. #13/2015) and the 2013 Helsinki Declaration. As regards the aMCI group, participants met the criteria for diagnosis of aMCI on the basis of the MCI working group of the European Consortium on AD [46]. The Mini-Mental State Examination (MMSE; [47-48]) mean score was 23.7 (SD= .71, range= 23-26) and the Clinical 5 Dementia rating scale (CDR; [49]) score was 0.5. Patients with multiple-domains or single non- memory domain MCI were not enrolled. As regards the mild probable AD group, patients met criteria of NINCDS-ADRDA for AD [15]. The mean MMSE score was 20.9 (SD= 1.9, range= 18-23) and the CDR score was 1 or higher. The NC group matched aMCI and eAD patients in terms of age, gender and education. NC were recruited from the public places via advertisements and by word of mouth (MMSE= 28.9, SD= .74, range= 27-30; CDR = 0). The three groups did not differ in age (F(2,35)=2.114, p> .05) and years of education (F(2,35)=1.698, p> .05). As expected, the three groups differed significantly on the MMSE score (F(2,35)=34.940, p< .001): the eAD group performed worse than both aMCI and NC groups; the aMCI group performed worse than NC (at least p< .05). of ro -p re Table 1. Demographic information about NC, aMCI and eAD patients. NC aMCI eAD lP Gender (F/M) 9/11 3/7 5/3 Age (range) 60-85 60-85 61-86 na Education, mean (sd) 13 (4.16) 12.5 (3.5) 10 (4.3) MMSE, mean (sd), range 28.9 (.74), 27-30 23.7 (.71), 23-26 20.9 (1.9), 18-23 ur Jo Neuropsychological assessment A general neuropsychological assessment was performed according to the Italian Neurological Society (SIN) guidelines for the diagnosis of dementia [50] and procedures of the Italian Interdisciplinary Network on AD (ITINAD) [51]. Besides the MMSE, patients underwent the Mental Deterioration Battery (MDB; [52]), the Frontal Assessment Battery (FAB; [53]), the Corsi Block-Tapping Test and the Three-Objects-Three-Places test, a test evaluating long-term visuospatial memory for the screening of AD [54]. NC achieved a normal score on these tests. By 6 contrast, AD patients showed a performance below the cut-off at the Three-Objects-Three-Places test (M= 6.4, SD=2.7; cut-off= 6–7). Participants were considered affected from aMCI if: i) satisfied Portet and colleagues’ clinical criteria [46]) exhibited pathological scores at immediate and/or delayed recall tasks of the MDB, with unimpaired performances at the remaining tests. Afterwards, all groups were submitted to the Ego-Allo/Cat-Coor task. Setting The experiment took place in a sound-proofed comfortable room of the Laboratories of Clinical Neuropsychology of “Ospedali dei Colli” (Naples, Italy). Participants were asked to sit on a chair in correspondence of the center of a small desk (100 x150 cm), 30 cm far from the edge of the desk. of ro Materials and Procedure Stimuli. The stimuli of the Ego-Allo/Cat-Coor task were the same as Iachini and colleagues in -p several previous studies [6; 35-36; 40-43]. They comprised six easily nameable actual geometrical objects (pyramid, parallelepiped, cone, cube, sphere, and cylinder). They varied in color re (dark/medium/light gray) and size: big objects (8x8 cm; except parallelepiped and cylinder: 8x11 cm) and small objects (6x6 cm; except parallelepiped and cylinder: 6x9 cm). By combining these lP features, 24 triads were obtained. To ensure that all triads were presented in the same way, 24 plasterboard panels (each one measuring 50x30 cm, 2 cm of thick) of the same size as the desk were used. On each panel, the shape forming the base of each object was engraved and the corresponding na object was placed there. Each triad was arranged on the corresponding panel according to the following criteria: (a) inter-object metric distances had to be easily distinguishable and (b) the metric distances were established in such a way that the amount of metric difficulty was the same ur for egocentric and allocentric judgments (see Figure 1 for an example). Participants were first given written instructions describing the experimental procedure, then there was a training session by Jo using three common objects (e.g., a glass, a cup, and a small box). Afterward, all experimental stimuli were presented and participants had to name them. In this way, difficulties due to naming problems could be excluded. Finally, the experiment started. 7 of Figure 1. The figure shows an example of how the triads could be placed. In the depicted case, the distances between stimuli were: cube–sphere= 11cm, sphere–cylinder= 17cm, cylinder–cube= ro 28cm. The cube and the cylinder were respectively 12 cm and 6 cm far from the body. The sphere was the target (T), that is the point of reference for the allocentric judgments. The metric difference -p between the two objects closer to the body was 6 cm (12–6 cm) and was the same as the metric difference between the two objects closer to the sphere (17–11 cm). The triads were presented at 30 cm from the edge of the desk just in front of participants. re lP Learning and Testing phase. Participants had to memorize (6 sec) the identity and positions of the three objects presented on the panel. Afterwards, they had to close their eyes while the experimenter removed the triad (5 sec). Then, the testing phase began (see Figure 2). Participants were asked to na verbally provide one of four types of spatial judgments (24 judgments for 24 triads): (a) egocentric- coordinate (Ego-Coor), “Which object was closest/farthest to you?”; (b) egocentric categorical (Ego-Cat), “Which object was on your left\right?”; (c) allocentric-coordinate (Allo-Coor), “Which ur object was closest/farthest to a target object (e.g., cilinder)?”; and (d) allocentric categorical (Allo- Cat), “Which object was on the left\right of the target object (e.g., cilinder)?”. For each judgment Jo accuracy (1 = correct; 0 = incorrect; score range = 0–6 for each spatial combination) and response time were manually recorded by the experimenter. Triads were associated to questions according to a balanced order and were randomly submitted. In this way, materials and order effects were controlled. 8 Figure 2. An example of trial is provided. From the left, participants saw the triad for 6 seconds, then the triad was removed and 5 seconds later the question was posed. Participants’ answers were not subject to time limits. Data analysis of A 2X2X2 ANOVAs for mixed designs was carried out on accuracy. “Groups” (NC vs eAD vs aMCI) was the three-level between variable, whereas “Reference Frames” (Egocentric vs ro Allocentric) and “Spatial Relations” (Categorical vs Coordinate) were the two within variables. The mean accuracy by each participant was calculated in terms of percentage of correct answers by -p dividing the total of correct answers by 6 (i.e. max accuracy for condition; e.g., Ego-Cat). The Tukey test was used to analyze post-hoc effects and the magnitude of the significant effects was re indicated by partial eta squared (η2p). Furthermore, in order to determine whether the task used in the current study allowed for a better lP classification of patients than existing tests, data from the Ego-Allo/Cat-Coor task were compared against data from the MMSE. To this aim: 1) A multivariate logistic regression was carried out with the dichotomous variable Normal na Controls (NC= 20) vs Patients (aMCI + AD =18) as criterion and the accuracy of the four spatial components and the MMSE scores as predictors; ur 2) To identify the specific contribution of each spatial component to the classification, univariate logistic regressions with the dichotomous variable NC vs Patients as criterion and the accuracy of Jo the four spatial components as predictors were performed; to make a direct comparison with the clinical measure, a univariate analysis was also carried out on MMSE; 3) Finally, to describe sensitivity and specificity of scores and to set the optimal cut-off level able to classify patients, the receiver operating characteristic (ROC) curve analysis was carried out. The ROC curve was created by plotting the score against specificity (X axis) and sensitivity (Y axis) assumed as gold standard, thus giving a measure of true-positive and true-negative rate. 9 RESULTS Accuracy. The ANOVA showed a main effect of “Groups” (F (1, 2)= 52.62, p< .001, ƞ2p= .75. Participants with eAD (M= .37, SD= .21) were less accurate than participants with aMCI (M= .69, SD= .25) and NC group (M= .88, SD= .15); aMCI participants were less accurate than NC participants (at least p < .001). Furthermore, a main effect of “Reference Frames” emerged (F (1, 2)= 21.47, p< .001, ƞ2p= .38). Allocentric judgments (M= .66, SD= .31) were less accurate than egocentric judgments (M= .78, SD= .22). An interaction effect between “Reference Frames” and “Spatial Relations” was also found (F (1, 2)= 4.9, p< .05, ƞ2p= .12). The interaction was due to Allo- Coor judgements (M= .61, SD= .30) being less accurate than all other judgements (Ego-Coor= .80, SD= .24; Ego-Cat = .77, SD= .21; Allo-Cat= .71, SD= .31, at least p< .05). Finally, a 3-way of interaction appeared: (F (1, 2)= 5.9, p< .01, ƞ2p= .25). In the NC group, no difference among the four judgments appeared (at least p> .12). In the aMCI group, Allo-Coor judgments were worse ro than Ego-Coor and Allo-Cat judgments (at least p< .05). In the eAD group, Allo-Cat judgments were worse than Ego-Cat judgements (p< .05). Table 2 shows descriptive statistics. As regards the -p comparison among the groups, results showed that eAD patients were worse than NC participants in all spatial judgments (at least p< .05). Moreover, eAD patients were worse than aMCI participants in Ego-Coor and Allo-Cat judgments (at least p< .05), but not statistically significant difference re appeared between eAD and aMCI for Ego-Cat (p> .05) and Allo-Coor (p> .05) judgments. Crucially, the only judgment in which aMCI participants performed worse than NC participants lP (p< .05), but no significant difference appeared as compared with the AD patients (p> .05), was the Allo-Coor one (see Figure 3). na ur Jo 10 of ro Figure 3. The figure shows the graphs for each spatial judgment: Ego-Cat = egocentric categorical, -p Ego-Coor = egocentric coordinate, Allo-Cat = allocentric categorical, Allo-Coor = allocentric coordinate. NC= normal control, aMCI= amnesic mild cognitive impairment, AD= Alzheimer re disease. The asterisk * indicates the presence of a significant difference with at least p< .05. The ≠ and = signs indicate the presence (at least p< .05) or the absence of a significant difference between lP groups, respectively. As an example, for the Ego-Cat judgments, there was no difference between NC and aMCI (=) and aMCI and AD (=), but a difference between NC and AD (*). na ur Jo Table 2. The average accuracy for each spatial judgement as a function of the participants’ group is reported. 95% CIs are indicated in parenthesis. 11 Groups Spatial judgments Ego-Cat Ego-Coor Allo-Cat Allo-Coor Mean Mean Mean Mean (-95%, + 95%) (-95%, + 95%) (-95%, + 95%) (-95%, + 95%) NC 0.90 0.94 0.87 0.80 (-0.83, +0.97) (-0.88, +1.01) (-0.79, +0.96) (-0.69, +0.90) aMCI 0.70 0.80 0.76 0.51 (-0.60, +0.79) (-0.71, +0.89) (-0.64, +0.88) (-0.37, +0.66) eAD 0.52 0.46 0.23 0.29 (-0.41, +0.62) (-0.35, +0.56) (-0.09, +0.36) (-0.13, +0.45) of Logistic regressions and ROC curves. The multivariate regression analysis model was significant: Chi²(5)=28.939 p< .001. This suggests that the four spatial components, together with the standard ro MMSE test, contributed significantly to the classification. The model gives rise to a percentage of correct classification = 89.47%. Considering the two groups, 19 out of 20 NCs were confirmed and, -p notably, 15 out of 18 aMCI/eAD individuals were classified as Patients. The univariate regression analysis on the MMSE scores showed a percentage of correct re classification = 88.89%, being the model significant (Chi²(1)=23.167 p< .001). In this case, 20 out of 20 NCs were confirmed, whereas only 14 out of 18 aMCI/eAD individuals were classified as lP Patients. The univariate analyses also showed that each spatial component gave a significant contribution to the classification. Specifically, the percentage of correct classification was 81.58% for egocentric categorical (Chi²(1)=20.053 p< .001), 83.34% for egocentric coordinate na (Chi²(1)=19.001 p< .001), 73.68% for allocentric categorical (Chi²(1)=13.901 p< .001), and 81.58% for allocentric coordinate judgments (Chi²(1)=19.728 p< .001). ur Finally, the areas under the ROC curve (AUC) of the four measures ranged from 0.810 (p< .001) for Allo-Cat to 0.874 (p< 0.001) for Ego-Cat values (see Table 3). The specificity was high for all measures, ranging from 85% (Ego-Cat) to 100% (Ego-Coor). Conversely, the sensitivity was weak Jo or modest for all but one measure, i.e. Ego-Cat, where a 77.78 % sensitivity was shown. The sensitivity shown by Ego-Cat was exactly the same observed for MMSE (i.e. 77.78 %) (see Figure 4). Table 3. The Area under the curve (AUC), sensitivity, specificity and cut-off values for each spatial judgement and for the MMSE are reported. 12 AUC (p) Sensitivity Specificity Cut-off Ego Cat 0.87 (0.001) 77.78 85.00 0.67 Ego Coor 0.83 (0.001) 66.67 100.00 0.66 Allo Cat 0.81 (0.001) 55.56 95.00 0.50 Allo Coor 0.86 (0.001) 66.67 95.00 0.50 MMSE 0.86 (0.001) 77.78 100.00 22.86 of ro -p re lP na ur Jo 13 of ro -p re lP na ur Jo Figure 4. The figure shows the graphs of the ROC curves of the MMSE (on the top) and four spatial measures, according to the best sensitivity and specificity assumed as gold standard. AUC = area under the curve; Ego-Cat = egocentric categorical; Ego-Coor = egocentric coordinate; Allo- 14 Cat= allocentric categorical; Allo-Coor= allocentric coordinate; MMSE= Mini-Mental State Examination. DISCUSSION In the past ten years, much evidence has suggested that reduced ability to encode, represent and retrieve from memory visual-spatial information, especially according to an allocentric reference frame, could be one of the early behavioral markers of the conversion from aMCI to Alzheimer's disease [among the others: 5-6; 9-10; 18]. Egocentric and allocentric reference frames combined with coordinate and categorical spatial relations are involved in mostly daily-life activities, from grasping objects to finding a destination. Clearly, alterations of the related underlying cerebral of structures can undermine the good functioning of these activities [1-3]. Therefore, a task that required the combination of coordinate-categorical relations with egocentric-allocentric frames was ro devised. The results showed that the overall performance of the eAD patients was worse than that of the NC -p and aMCI participants, whereas the aMCI participants performed worse than the NC participants [e.g. 2-3; 18]. In line with previous literature, a main effect of reference frames emerged with re allocentric being less accurate than egocentric judgments. Moreover, a two way interaction showed that allocentric coordinate judgments were less accurate than all the other spatial judgments [34-35; lP 55-57]. However, these effects were qualified by a significant three way interaction due to the different groups. Indeed, in the aMCI group a specific difficulty with allocentric coordinate judgments appeared: these judgments were worse than egocentric coordinate and allocentric na categorical judgments. In the eAD group, a general difficulty with the allocentric components emerged. Instead, in the healthy elderly group there appeared no difference among the four judgments. Importantly, the comparison between the different groups revealed a novel finding: the ur only combination in which the aMCI participants performed worse that the NC participants, but their performance did not statistically differ from that of the eAD patients, was the allocentric Jo coordinate one. Moreover, the performance of aMCI participants was better than that of eAD patients in the allocentric categorical and egocentric coordinate combinations. As regards the egocentric categorical combination, the performance of aMCI participants did not statistically differ from either NC or eAD participants. This suggests that aMCI participants, as compared to NC participants, started to have a specific difficulty in processing metric distances according to external, not body-related, elements. 15 Notably, the decline observed in the allocentric coordinate judgments is substantially in line with studies adopting spatial navigational tasks [e.g. 16; 21]. For example, in Laczó and colleagues’ study [21], both aMCI and AD patients were submitted to an adapted-for version of the Morris Water Maze Task. In this task, participants had to learn a route to reach a specific hidden position within the maze. To this aim they could use an egocentric or an allocentric strategy. The allocentric strategy was based on distances between the hidden positions and cues in the environment, i.e. allocentric coordinate information. Results showed that both aMCI and AD patients failed to adopt an allocentric strategy [see also 3; 6; 16; 58-60]. However, no study so far has explicitly verified if this allocentric difficulty is linked to metric relations, as supported by the current study, or to more abstract/verbal spatial categories. of The specific allocentric coordinate difficulty in aMCI could have different but no exclusive explanations. On the one hand, coordinate relationships are more difficult to process than ro categorical relationships because they require fine-grained inspection of distances that vary from one object to another (for a review, [27]). Therefore, they are more resources demanding than -p categorical judgments where the same verbal category (e.g. right/left or above/below) is attributed to a group of elements in the space. On the other hand, it is possible that this difficulty is due to a general activity decrease in the right hemisphere during normal aging, as suggested by the right re hemi-aging hypothesis (for a review, [61]). This hypothesis was based on the evidence that spatial task performance tends to decline more rapidly than verbal task performance as age increases [62; lP 36], and on the idea that some structures in the right hemisphere might undergo a faster aging process than the corresponding structures in the left hemisphere [63; 37]. More importantly, the activity decrease on the right hemisphere would become even more marked in the aMCI than in na healthy aging [39]. Finally, it is possible that the aMCI and AD patients’ performance observed in this study is related to dysfunction in some brain areas responsible of path integration [64-65]. Path ur integration is the ability to use self-motion cues to estimate the distance and direction to a point of origin, and relies also on the ability to use allocentric representations [66; 4]. Consistent with studies that have shown a strict relationship between hippocampus and allocentric representations, Jo Tu and colleagues [64] have shown a link between hippocampal atrophy and poor allocentric performance in a path integration task of patients with frontotemporal dementia and AD. Moreover, Howett and colleagues [65] found that MCI biomarker-positive patients, i.e. those with prodromal Alzheimer's disease, made significantly larger absolute distance errors (i.e. metric/coordinate errors), in a path integration task, compared to MCI patients with negative biomarkers. More interestingly, the authors trace back errors in the distance estimation task to reduced volumes of the posteromedial part of the entorhinal cortex (pmEC). 16 In terms of egocentric performance, AD patients, unlike aMCI participants, showed a general difficulty compared to healthy elderly people. The reduced egocentric ability of AD patients could probably reflect the involvement of extra-hippocampal alterations. In the AD group, these alterations could compromise the egocentric compensation strategies used by NC [2; 57; 67] and aMCI. In sum, the allocentric alterations associated with hippocampal atrophy would precede the egocentric impairments, and this could reflect the neuropathological distribution progressively involving the medial temporal lobe, the postero-medial areas and then the associative and temporo- parietal areas [2-3; 5; 18-19]. The specific decline in allocentric coordinate judgments of aMCI of individuals could represent an early indicator of detachment from a healthy aging condition to Alzheimer's disease. This is important not only at theoretical level but also from a clinical point of ro view. In fact, it suggests that an allocentric deficit is not always clinically detectable in aMCI individuals as it may be mitigated when the reference frame is not combined with metric properties. -p One might argue that spatial abilities assessment in a small-scale environment may be weakly related to the navigational difficulties in everyday large scale environments encountered by patients re at risk of developing Alzheimer’s disease. The main difference between small and large scale environmental learning is that, in the latter, spatial information is not entirely and immediately lP visible at a glance, but it results from the integration of visual and kinesthetic information [68-69]. For this reason, nowadays many studies assess aMCI and Alzheimer’s patients’ visual-spatial difficulties in large-scale virtual environments [3; 70; 71]. However, the exploration of these virtual na environments requires enormous cognitive efforts both in terms of working memory processes and attentional resources, making it difficult to identify the nature of the navigational problems (but see [65]). Besides, most of the daily navigational issues that patients encounter are usually collected on ur the basis of personal accounts or narrated by family members. For this reason, if on the one hand the task used in our study does not reflect the multifactorial nature of navigation, on the other hand Jo it has the advantage of assessing different types of basic spatial components and their combinations that could be involved in different spatial tasks. The clinical interview with aMCI patients, as recently suggested by Coughlan and colleagues [72], should always be accompanied by a standardized and validated spatial diagnostic test battery that does not rely on topographical memory, but that allows to investigate specific cognitive factors involved in navigation. In this regard, the Ego-Allo/Cat-Coor task used in this study could integrate existing assessment protocols to detect the early stages of conversion from pre-dementia stage to 17 AD [73]. Indeed, results from the logistic and ROC analyses have shown that each spatial component (i.e. Ego-Coor, Ego-Cat, Allo-Coor, Allo-Cat), together with the MMSE, gave a significant contribution to the classification of the patients as aMCI or AD. In fact, the analysis confirmed that 7 out of 10 aMCI persons were classified as patients and this classification was very close to the data from follow-up showing that 8 out of 10 aMCI individuals converted into AD. In particular, the specificity was high for all measures, ranging from 85% (Ego-Cat) to 100% (Ego- Coor). This means that the used measures have a high positive predictive value that allows to easily identify individuals with a “negativity in health” [74]. Conversely, the sensitivity, that is the capacity to exclude healthy individuals from the clinical population, was weak or modest for all but one measures, i.e. Ego-Cat, where a 77.8 % sensitivity was shown. Interestingly, the sensitivity of shown by Ego-Cat was exactly the same observed for the MMSE. MMSE also showed good properties in discriminating patients (aMCI and AD) well from NC. However, the optimal cut-off score corresponding to the better sensitivity/specificity (25.86 score) was consistently higher than ro the cut-off values reported in all normative Italian studies [75-77]. With all due caution due to the small sample size, this result might point to a limitation on the use of this tool in clinical settings, -p given the possible overlapping, in the general population, between normal subjects and cognitively impaired patients, in particular if affected from aMCI. re In conclusion, the results from the current study showed a specific difficulty in aMCI and eAD in processing allocentric coordinate spatial information. Furthermore, they suggest that the Ego- lP Allo/Cat-Coor spatial memory task may usefully complement the standard assessment protocols used in clinical settings. However, we acknowledge that the small sample size represents a limitation, but the results are promising and deserve future investigations with larger clinical na samples. ur Author statement Gennaro Ruggiero (Conceptualization; Formal analysis; Investigation; Methodology; Writing – Jo original draft; Writing – review & editing) Francesco Ruotolo (Data curation; Investigation; Formal analysis; Writing – review & editing) Alessandro Iavarone (Resources; Methodology; Data curation; review & editing) Tina Iachini (Data curation; Supervision; Writing – review & editing) CONFLICT OF INTEREST The authors declare no conflict of interest. 18 Jo ur na lP re -p ro of 19 REFERENCES [1] Aguirre GK, D'Esposito M. Topographical disorientation: A synthesis and taxonomy, Brain 1999; 122: 1613-28. [2] Lithfous S, Dufour A, Despres O. Spatial navigation in normal aging and the prodromal stage of Alzheimer's disease: Insights from imaging and behavioral studies. Ageing Res Rev 2013; 12: 201- 13. [3] Moffat SD. Aging and spatial navigation: What do we know and where do we go? Neuropsychol Rev 2009; 19: 478-89. [4] Burgess N, Trinkler I, King J, Kennedy A, Cipollotti L. Impaired allocentric spatial memory of underlying topographical disorientation. Rev Neuroscience 2006; 17:239 –251. doi:10.1515/ REVNEURO.2006.17.1-2.239 ro [5] Iachini I, Iavarone A, Senese VP, Ruotolo F, Ruggiero G. Visuospatial memory in healthy elderly, AD and MCI: A review. Curr Aging Sci 2009; 2: 43-59. -p [6] Ruggiero G, Iavarone A, Iachini T. Allocentric to Egocentric Spatial Switching: Impairment in aMCI and Alzheimer's Disease Patients? Curr Alzheim Rese 2018, 15(3):229-236. doi: re 10.2174/1567205014666171030114821. [7] Vann SD, Aggleton JP, Maguire EA. What does the retrosplenial cortex do? Nat Rev Neurosci lP 2009; 10: 792-802. [8] Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern Med 2004; 256: 183–94. na [9] Laukka EJ, MacDonald SW, Fratiglioni L, Bäckman L. Preclinical cognitive trajectories differ for Alzheimer's disease and vascular dementia. J Int Neuropsych Soc 2012; 18(2): 191-199. [10] Mortamais M, Ash JA, Harrison J, Kaye J, Kramer J, Randolph C, ... & Ritchie K. Detecting ur cognitive changes in preclinical Alzheimer's disease: A review of its feasibility. Alzheimer's & Dementia 2017, 13(4): 468-492. Jo [11] deIpolyi A, Rankin K, Mucke L, Miller B, Gorno-Tempini M. Spatial cognition and the human navigation network in AD and MCI. Neurology 2007; 69: 986-97. [12] Hort J, Laczó J, Vyhnálek M, Bojar M, Bures J, Vlček K. Spatial navigation deficit in amnestic mild cognitive impairment. Proc Natl Acad Sci USA 2007; 10: 4042-7. 20 [13] Kalová E, Vlcek K, Jarolimova E, Bure J. Allothetic orientation and sequential ordering of places is impaired in early stages of Alzheimer's disease: Corresponding results in real space tests and computer tests. Behav Brain Res 2005; 159: 175-86. [14] Kessels RP, Feijen J, Postma A. Implicit and explicit memory for spatial information in Alzheimer’s disease. Dement Geriatr Cogn Disord 2005; 20: 184–91. [15] McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack Jr CR, Kawas CH, et al. The diagnosis of dementia due to Alzheimer's disease: Recommendations from the National Institute on Aging and the Alzheimer's Association workgroup. Alzheimers Dement 2011; 7: 263-9. [16] Nedelska Z, Andel R, Laczò J, Vlček K, Horinek D, Lisy J, et al. Spatial navigation impairment is proportional to right hippocampal volume. P Natl Acad SciBiol 2012; 109: 2590-4. of [17] Vlček K, Laczó J. Neural correlates of spatial navigation changes in mild cognitive impairment ro and Alzheimer’s disease. Front Behav Neurosci 2014; 8:1-6. [18] Serino S, Cipresso P, Morganti F, Riva G. The role of egocentric and allocentric abilities in -p Alzheimer's disease: A systematic review. Ageing Res Rev 2014; 16: 32-44. [19] Albert MS, DeKosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, et al. The diagnosis of re mild cognitive impairment due to Alzheimer's disease: Recommendations from the National Institute on Aging and Alzheimer's Association workgroup. Alzheimers Dement 2011 ; 7: 270-9. lP [20] Dubois B, Albert ML. Amnestic MCI or prodromal Alzheimer’s disease? Lancet Neurol 2004; 3: 246–8. [21] Laczò J, Vlček K, Vyhna´lek M, Vajnerova O, Ort M, Holmerova I, et al. Spatial navigation na testing discriminates two types of amnestic mild cognitive impairment. Behav Brain Res 2009; 202: 252-9. ur [22] Laczò J, Andel R, Vlček K, Vyhna´lek M, Magerová H, Varjassyova A, et al. Spatial navigation deficits predict Alzheimer's disease in patients with amnestic mild cognitive impairment: Jo A 3-year follow-up study. Eur J Neurol 2010; 17: 44. [23] Laczò J, Andel R, Vlček K, Maťoška V, Vyhna´lek M, Tolar M, et al. Spatial Navigation and APOE in Amnestic Mild Cognitive Impairment. Neurodegener Dis 2011; 8: 169-77. [24] Laczò J, Andel R, Vyhna´lek M, Vlček K, Magerová H, Varjassyova A, et al. From Morris Water Maze to Computer Tests in the Prediction of Alzheimer's Disease. Neurodegener Dis 2012; 10: 153-7. 21 [25] Mapstone M, Steffenella TM, Duffy CJ. A visuospatial variant of mild cognitive impairment: Getting lost between aging and AD. Neurology 2003; 60: 802–8. [26] Plancher G, Tirard A, Gyselinck V, Nicolas S, Piolino P. Using virtual reality to characterize episodic memory profiles in amnestic mild cognitive impairment and Alzheimer's disease: Influence of active and passive encoding. Neuropsychologia 2012; 50: 592-602. [27] Jager G and Postma A. On the hemispheric specialization for categorical and coordinate spatial relations: a review of the current evidence. Neuropsychologia 2003; 41(4): 504-515. [28] Kosslyn SM, Ed. Image and brain: The resolution of the imagery debate. Cambridge, MA, US, The MIT Press (1994). of [29] Ruotolo F, Iachini T, Postma A, van Der Ham IJM. Frames of reference and categorical and coordinate spatial relations: A hierarchical organisation. Exp Brain Res 2011a; 214: 587-95. ro [30] Ruotolo F, van Der Ham IJM, Iachini T, Postma A. The relationship between allocentric and egocentric frames of reference and categorical and coordinate spatial information processing. Quarterly J Exp Psychol 2011b; 64: 1138–56. -p [31] Milner AD, Goodale MA. Two visual systems re-viewed. Neuropsychologia 2008; 46:774-85. re [32] Ruotolo F, Ruggiero G, Raemaekers M, Iachini T, van der Ham IJM, Fracasso A, Postma, A. Neural correlates of egocentric and allocentric frames of reference combined with metric and non- lP metric spatial relations. Neuroscience 2019, 409:235-252. [33] Neggers SFW, van der Lubbe RHJ, Ramsey NF, Postma A. Interactions between ego- and allocentric neuronal representations of space. NeuroImage 2006; 31:320–31. na [34] Ruotolo F, Iachini T, Ruggiero G, van der Ham IJM, Postma A. Frames of reference and categorical/coordinate spatial relations in a “what was where” task. Exp Brain Res 2016. ur doi:10.1007/s00221-016-4672-y [35] Ruotolo F, van der Ham I, Postma A, Ruggiero G, Iachini T. How coordinate and categorical Jo spatial relations combine with egocentric and allocentric reference frames in a motor task: Effects of delay and stimuli characteristics. Behav Brain Res 2015; 284: 167–78. [36] Bruyer R, Scailquin J-C, Coibion, P. Dissociation between categorical and coordinate spatial computations: Modulation by the cerebral hemispheres, task properties, mode of response, and age. Brain Cognition, 1997; 33:245–277. [37] Antonova E, Parslow D, Brammer, M, Dawson, GR, Jackson, SHD, Morris R G. Age-related neural activity during allocentric spatial memory. Memory 2009, 17(2): 125-143. 22 [38] van der Ham IJ, van Wezel RJ, Oleksiak A, van Zandvoort MJ, Frijns CJ, Kappelle LJ, Postma A. The effect of stimulus features on working memory of categorical and coordinate spatial relations in patients with unilateral brain damage. Cortex 2012; 48(6):737-745. [39] Hampstead BM, Stringer AY, Stilla RF, Amaraneni A, Sathian K. Where did I put that? Patients with amnestic mild cognitive impairment demonstrate widespread reductions in activity during the encoding of ecologically relevant object-location associations. Neuropsychologia 2011; 49(9): 2349-2361. [40] Iachini T, Ruggiero G. Egocentric and allocentric spatial frames of reference: A direct measure. Cogn Process 2006; 7: 126–7. [41] Iachini T, Ruggiero G, Conson M, Trojano L. Lateralization of egocentric and allocentric of spatial processing after parietal brain lesions. Brain Cognition 2009b; 69: 514–20. ro [42] Iachini T, Ruggiero G, Ruotolo F. Does blindness affect egocentric and allocentric frames of reference in small and large scale spaces? Behav Brain Res 2014; 273: 73-81. -p [43] Ruggiero G, Ruotolo F, Iachini T. Egocentric/allocentric and coordinate/categorical haptic encoding in blind people. Cogn Process 2012; 13: 313–7. re [44] Ruggiero G, Ruotolo F, Iachini T. The role of vision in egocentric and allocentric spatial frames of reference. Cogn Process 2009; 10: 283–5. lP [45] Committeri G, Galati G, Paradis AL, Pizzamiglio L, Berthoz A and LeBihan D. Reference frames for spatial cognition: different brain areas are involved in viewer-, object-, and landmark- centered judgments about object location. J Cogn Neurosci 2004; 16(9): 1517-1535. na [46] Portet F, Ousset PJ, Visser PJ, et al. Mild cognitive impairment (MCI) in medical practice: A critical review of the concept and new diagnostic procedure. Report of the MCI Working Group of the European Consortium on Alzheimer’s Disease. J Neurol Neurosurg Psychiatry 2006; 77: 714-8. ur [47] Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical method for grading Jo the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: 189-98. [48] Measso G, Cavarzeran F, Zappalà G, Lebowitz BD, Crook TH, Pirozzolo FJ, et al. The Mini Mental State Examination: Normative study of an Italian random sample. Dev Neuropsychol 1993; 9: 77-85. [49] Morris JC. The Clinical Dementia Rating (CDR): Current version and scoring rules. Neurology 1993; 43: 2412-14. 23 [50] Musicco M, Caltagirone C, Sorbi S, Bonavita V. Italian Neurological Society guidelines for the diagnosis of dementia: Revision I. Neurol Sci 2004; 25: 154–82. [51] Perri R, Serra L, Carlesimo GA, Caltagirone C. Amnestic mild cognitive impairment: Difference of memory profile in subjects who converted or did not convert to Alzheimer’s disease. Neuropsychology 2007; 21: 549–558. [52] Carlesimo GA, Caltagirone C, Gainotti G. The Mental Deterioration Battery: Normative data, diagnostic reliability and qualitative analyses of cognitive impairment. The Group for the Standardization of the Mental Deterioration Battery. Eur Neurol 1996; 36: 378–84. [53] Iavarone A, Ronga B, Pellegrino L, Loré E, Vitaliano S, Galeone F. The Frontal Assessment Battery (FAB): Normative data from an Italian sample and performances of patients with of Alzheimer’s disease and frontotemporal dementia. Funct Neurol 2004; 19: 191-5. ro [54] Prestia A, Rossi R, Geroldi C, Galluzzi S, Ettori M, Alaimo G, Frisoni GB. Validation study of the Three-Objects-Three-Places Test: a screening test for Alzheimer's disease. Experimental aging research 2006, 32(4):395-410. -p [55] Ruggiero G, Ruotolo F, Orti R, Rauso B, Iachini T. Egocentric metric representations in re peripersonal space: A bridge between motor resources and spatial memory. British Journal of Psychology (under review). lP [56] Ruggiero G, Frassinetti F, Iavarone A, Iachini T. The lost ability to find the way: Topographical disorientation after a left brain lesion. Neuropsychology 2014; 28: 147-60. [57] Ruggiero G, D’Errico O, Iachini T. Development of egocentric and allocentric spatial na representations from childhood to elderly age. Psychol Res 2016; 80: 259-72. [58] Caffò AO, De Caro MF, Picucci L, Notarnicola A, Settanni A, Livrea P, et al. Reorientation ur deficits are associated with amnestic mild cognitive impairment. Am J Alzheimers Dis Other Demen 2012; 27: 321-30. Jo [59] Caffò AO, Hoogeveen F, Groenendaal M, Perilli AV, Picucci L, Lancioni GE, et al. Intervention strategies for spatial orientation disorders in dementia: A selective review. Dev Neurorehabil 2014; 17. http://dx.doi.org/10.3109/17518423.2012.749951 [60] Lester AW, Moffat SD, Wiener JM, Barnes CA, Wolbers T. The aging navigational system. Neuron 2017; 95(5):1019-1035. [61] Dolcos F, Rice HJ, Cabeza R. Hemispheric asymmetry and aging: right hemisphere decline or asymmetry reduction. Neurosci Biobehav Rev 2002; 26(7): 819-825. 24 [62] Goldstein G, Shelly C. Does the right hemisphere age more rapidly than the left?. J Clin Exp Neuropsychol 1981; 3(1): 65-78. [63] Gerhardstein P, Peterson MA, Rapcsak SZ. Age-related hemispheric asymmetry in object discrimination. J Clin Exp Neuropsychol 1998; 20(2): 174-185. [64] Tu S, Spiers HJ, Hodges JR, Piguet O, Hornberger M. Egocentric versus allocentric spatial memory in behavioral variant frontotemporal dementia and Alzheimer’s disease. J Alzheimers Dis 2017; 59(3): 883-892. [65] Howett D, Castegnaro A, Krzywicka K, Hagman J, Marchment D, Henson R, Rio M, King JA, Burgess N, Chan D. Differentiation of mild cognitive impairment using an entorhinal cortex-based test of virtual reality navigation. Brain 2019; 142(6): 1751-1766. of [66] Etienne AS, Jeffery KJ. Path integration in mammals. Hippocampus 2004; 14(2): 180-192. ro [67] Rodgers MK, Sindone JAIII, Moffat SD. Effects of age on navigation strategy. Neurobiol Aging 2012; 33: 15-22. -p [68] Hegarty M, Montello DR, Richardson AE, Ishikawa T, Lovelace K. Spatial abilities at different scales: Individual differences in aptitude-test performance and spatial-layout learning. re Intelligence 2006; 34(2): 151-176. [69] Iachini T, Ruggiero G, Ruotolo F. Does blindness affect egocentric and allocentric frames of lP reference in small and large scale spaces?. Behav Brain Res 2014; 273: 73-81. [70] Cogné M, Taillade M, N’Kaoua B, Tarruella A, Klinger E, Larrue F, Sauzéonac H, Joseph PA, Sorita E. The contribution of virtual reality to the diagnosis of spatial navigation disorders and to na the study of the role of navigational aids: A systematic literature review. Ann Phys Rehabil Med 2017; 60(3): 164-176. ur [71] Colombo D, Serino S, Tuena C, Pedroli E, Dakanalis A, Cipresso P, Riva G. Egocentric and allocentric spatial reference frames in aging: A systematic review. Neurosci Biobehav Rev 2017; Jo 80: 605-621. [72] Coughlan G, Laczó J, Hort J, Minihane AM, Hornberger M. Spatial navigation deficits— overlooked cognitive marker for preclinical Alzheimer disease?. Nat Rev Neurol 2018; 14(8): 496- 506. [73] Gainotti G, Quaranta D, Vita MG, Marra C. Neuropsychological Predictors of Conversion from Mild Cognitive Impairment to Alzheimer's Disease. J Alzheimer Dis 2014; 38: 481-95. 25 [74] Florkowski C M. Sensitivity, specificity, receiver-operating characteristic (ROC) curves and likelihood ratios: communicating the performance of diagnostic tests. Clin Biochem Rev 2008; 29(Suppl 1): S83. [75] Measso G, Cavarzeran F, Zappalà G, et al. The mini‐ mental state examination: Normative study of an Italian random sample. Dev Neuropsychol 1993; 9: 77-85 [76] Magni E, Binetti G, Bianchetti A, et al. Mini-Mental State Examination: A Normative Study in Italian Elderly Population. Eur J Neurol 1996; 3: 198-202 [77] Carpinelli Mazzi M, Iavarone A, Russo G, et al. Mini-Mental State Examination: new normative values on subjects in Southern Italy. Aging Clin Exper Res 2020 32:699-702. of ro -p re lP na ur Jo 26