Abstract
This study investigates phenomenological reliving of future thinking in Alzheimer’s disease (AD) patients and matched controls. All participants were asked to imagine in detail a future event, and afterward, were asked to rate phenomenological characteristics of their future thinking. As compared to controls, AD participants showed poor rating for reliving, travel in time, visual imagery, auditory imagery, language, and spatiotemporal specificity. However, no significant differences were observed between both groups in emotion and importance of future thinking. Results also showed lower rating for visual imagery relative to remaining phenomenological features in AD participants compared to controls; conversely, these participants showed higher ratings for emotion and importance of future thinking. AD seems to compromise some phenomenological characteristics of future thinking, especially, visual imagery; however, other phenomenological characteristics, such as emotion, seem to be relatively preserved in these populations. By highlighting the phenomenological experience of future thinking in AD, our paper opens a unique window into the conscious experience of the future in AD patients.
INTRODUCTION
Humans have an extraordinary ability to disengage from the immediate environment in order to contemplate hypothetical future scenarios [1, 2]. This capacity to project oneself into the future (i.e., future thinking) has a significant adaptive value as it allows humans to evaluate potential consequences prior to acting [3, 4]. Besides being a core element for decision making [3], future thinking has been found to be important for emotion regulation [5] as well as to behavioral flexibility and effective planning while achieving goals [6–8]. Future thinking has also been found to be intimately linked to episodic memory. According to the constructive episodic simulation hypothesis, future thinking requires retrieving detailed information from episodic memory and flexibly recombining them into a mental simulation of the future [7, 9]. Based on the constructive episodic simulation hypothesis, the compromise of future thinking has been observed in patients with episodic memory decline, especially in those with Alzheimer’s disease (AD) [10].
In a pioneering study on future thinking, Addis et al. [11] asked AD participants to remember past events and to simulate future events. For each event, participants rated emotional intensity, and personal significance (e.g., how important the event). AD participants generated few specific details in both past and future thinking tests. Interestingly, no significant differences were observed between past and future thinking in regard to emotional intensity or personal significance. The compromise of specificity in both past and future thinking in AD, as observed in the study by Addis et al. [11], was also observed in a study by Irish and colleagues [12]. The similarity between past and future thinking in AD reported in the study by Addis et al. [11] was also confirmed in another study that showed similar autobiographical specificity in past and future thinking in AD [13]. In the latter study, AD participants with low cognitive flexibility showed higher similarity of past and future thinking than those with high flexibility; these findings were attributed to deficits in the frontal lobes and hippocampus, which compromise the ability to recombine episodic information into novel and flexible configurations as scenarios for the future. Future thinking in AD was investigated in another study [14] that analyzed past and future projection in AD participants and controls with respect to specificity, ability to remember contextual information (e.g., when and where, and who), and personal significance. In addition, participants were asked to evaluate their subjective reliving in past and future thinking by providing “relive/re-experience” responses if they recollected subjective experiences from the encoding context or a “recognize” response if retrieval did not trigger such a recollection. Results showed that past and future thinking in AD participants triggered similar autobiographical specificity, similar personal significance, and similar autonoetic states. Interestingly, fewer “Remember” responses were provided by AD participants than by controls during both past and future thinking, suggesting a compromise of phenomenological reliving in past and future projection in AD. Based on the latter findings, the main aim of the present paper is to provide a fine-grained dissection of the phenomenological reliving during future thinking in AD.
Since the main aim of the present paper is to provide a fine-grained dissection of phenomenological reliving during future thinking in AD, it is important to highlight a prior study assessing this aim for past thinking. In another study [15], AD participants and controls retrieved past events, so as to rate reliving (“I feel as though I am reliving the original event”), back in time (“I feel that I travel back to the time it happened”), remembering (“I can actually remember it rather than just knowing that it happened”), how memories seem real (“I believe the event in my memory really occurred in the way I remember it”), visual imagery (“I can see it in my mind”), auditory imagery (“I can hear it in my mind”), language (“I or other people are talking”), emotion (“I can feel now the emotions I felt then”), importance (“this memory is significant for my life”), rehearsal (“since it happened, I have thought or talked about this event”), and spatiotemporal specificity (“I can recall the place/time where it occurred”). As compared to controls, AD participants showed lower rating of reliving, travel in time, remembering, realness, visual imagery, auditory imagery, language, rehearsal, and spatiotemporal specificity— a reduction that was especially pronounced for visual imagery. However, AD participants showed high rating for emotion and importance. El Haj et al. [15] suggested that AD might compromise many phenomenological features of past thinking, especially visual imagery, but also might preserve some other features. The present study assesses the same measures but in relation to future thinking.
The evaluation of AD effects on phenomenological reliving of future thinking can benefit from studies assessing the same aim in young healthy populations. These studies evaluated a variety of phenomenological characteristics by asking participants to imagine future events and to rate these events according to overall vividness, clarity of location, clarity of sounds, smell/tastes, emotions, feelings of mentally traveling backward or forward in time, or degree to which the event is remembered or imagined as a coherent story [8, 16–19]. In our study, these ratings are simplified to fit the limited cognitive abilities of AD patients.
Together, research has suggested a compromise of future thinking in AD [11–14]. Our paper extended this research by assessing the effect of AD on phenomenological reliving of future thinking. Prior research has assessed phenomenological reliving only in regard to emotion and personal significance [11], or in regard to general reliving [14], dimensions that provide a general rather than a fine-grained view of phenomenological reliving. Because phenomenological reliving includes a wide variety of dimensions such as visual imagery, mental imagery, feelings of mentally traveling in time, coherence, and spatiotemporal specificity [20], it would be of interest to investigate all these dimensions for future thinking in AD. Thanks to this fine-grained approach, we may understand which phenomenological features are compromised by or preserved in AD this in light of research suggesting a selective compromise for phenomenological features of past thinking [15].
METHODS
Participants
We tested 28 participants with a clinical diagnosis of probable mild AD (19 women and 9 men; M age = 72.50 years, SD = 6.79; M years of formal education = 8.79, SD = 2.62) and 31 control older adults (20 women and 11 men; M age = 72.81 years, SD = 7.75; M years of formal education = 9.94, SD = 3.01). The AD participants were recruited from local retirement homes and the diagnosis (i.e., probable AD) was made by an experienced neurologist or geriatrician according to the criteria developed by the National Institute on Aging and the Alzheimer’s Association criteria for probable Alzheimer’s disease [21]. The control participants were often spouses or companions of AD participants, and were independent and living at home. These participants were matched with the AD participants according to age [t(57) = 0.16, p > 0.10], sex [χ2 (1, N = 59) = 0.73, p > 0.10], and educational level [t(57) = 1.53, p > 0.10].
Exclusion criteria for all participants were significant neurological or psychiatric illness and alcohol or drug use. None of the participants presented any major visual or auditory acuity difficulties that could prevent assessment. They freely consented to participate and were able to withdraw whenever they wished. Cognitive characteristics of participants were assessed with a comprehensive battery detailed below.
Cognitive characteristics
Cognitive characteristics of all participants were evaluated with a battery tapping general cognitive functioning, episodic memory, working memory, inhibition and depression. Scores are summarized in Table 1.
Cognitive characteristics of Alzheimer’s disease (AD) patients and control participants
Note. Standard deviations are given between brackets; the maximum MMSE score was 30 points; the maximum score on the Grober and Buschke task was 16 points; performances on the forward and backward spans refer to number of correctly repeated digits; score on the Stroop task refers to reaction time; the cut-off on the HADS (Hospital Anxiety and Depression Scale) was >10/2 points; differences between groups were significant at: **p < 0.01, ***p < 0.001; after checking for normality of distributions, comparisons were made with Student’s t-test; effect size was calculated by using Cohen’s d criterion [34] (0.20 = small, 0.50 = medium, 0.80 = large).
General cognitive functioning
We used the Mini-Mental State Examination [22] and the maximum score was 30 points.
Episodic memory
We used a French version [23] of the episodic task of Grober and Buschke [24] in which the participants had to retain 16 words, each describing an item belonging to a different semantic category. Immediate cued recall was succeeded by a distraction phase during which participants had to count backwards from 374 in 20 s. This distraction phase was succeeded by 2 min of free recall and the score from this phase provided a measure of episodic recall (16 points maximum).
Working memory
Participants were asked to repeat a string of single digits in the same order (i.e., forward spans) or in reverse order (i.e., backward spans).
Inhibition
We used the Stroop task where the score referred to completion time for the interference condition minus the average completion time for the word reading and color naming conditions.
Depression
We used the Hospital Anxiety and Depression Scale [25] which consists of seven items that were scored by participants on a four-point scale ranging from 0 (not present) to 3 (considerable). The cut-off for definite depression was set at >10/21 points [26].
Procedures
Past and future thinking
Participants were tested individually in their homes or their rooms/apartments (in retirement homes). They were informed that they were taking part in a study examining their cognitive performance as assessed with the neuropsychological battery. Assessment occurred in two sessions, counterbalanced and separated by approximately one week. One session included past thinking and the other future thinking. In each session, participants were asked to “recount in detail an event in their lives” or “imagine in detail a future event”, regardless of when the event occurred or will occur. When constructing future events, participants were instructed to imagine events that might reasonably happen in the future. Participants were also instructed not to describe a past event or any part of it, but rather to imagine something completely new. For past and future thinking, the investigator explained that participants had be precise and specific: events had to have lasted/last no more than a day and details had to be provided, such as time and place at which events had/will have occurred. Some examples were provided to illustrate what would be considered as a specific event. Participants were also asked to describe their feelings and emotions during these events. Participants were allowed three minutes, and the duration was stated from the onset so participants could plan accordingly. This time limit was adopted to avoid potential redundancy and/or distractibility [11, 28] and was kept constant to control for duration of sustained mental effort. Past and future thinking were recorded using a smartphone and were transcribed at a later time.
Phenomenological evaluation of past and future thinking
After memory generation, we assessed phenomenological reliving by asking participants to rate on a five-point scale (zero points = not at all, one point = slightly, two points = moderately, three points = quite a bit, and four points = extremely) 1) reliving “I feel as though I am re/pre experiencing that event”, 2) travel in time “I feel that I travel back to the time it happened/will happen”, 3) significance “this event is/will be significant for my life”, 4) visual imagery “I can see it in my mind”, 5) auditory imagery “I can hear it in my mind”, 6) language “I or other people are talking”, 7) emotion “I can feel now the emotions I felt/will feel then”, 8) spatial specificity “I can recall/imagine the place where it occurred/will occur”, and 9) temporal specificity “I can recall/imagine the time when it occurred/will occur”. These nine dimensions were assessed in light of the Autobiographical Memory Questionnaire [20] that deals with phenomenological reliving of past thinking, as well as in light of research assessing phenomenological reliving of future thinking in normal populations [8, 16–19].
Since some AD participants had difficulties with phenomenological ratings, further instructions were provided. For instance, when rating reliving and back in time, participants were informed that the fourth point on the scale corresponded to “as clearly as if the event was happening now”. Participants were also presented with printed scales and statements (Time New Roman 48) so they could read them when rating.
Evaluation of autobiographical specificity
The subjective experience, as rated by the participants, was enriched by an objective evaluation of autobiographical specificity. Specificity of both past and future thinking was scored with the TEMPau scale (Test épisodique de mémoire du passé, [29–31]), an autobiographical evaluation instrument inspired by classic autobiographical evaluations [32]. For each event, we gave zero points if there was no memory or only general information about a theme; one point for a repeated or an extended event; two points for an event situated in time or/and space; three points for a specific event lasting less than 24 h and situated in time and space; and four points for a specific event situated in time and space enriched with phenomenological details. Thus, the maximum specificity score for each past or future event was four points. To avoid bias in scoring, events were rated by the first author and an independent rater who was blind to the study objectives and to individual participants’ group membership (AD patients vs. controls). Using Cohen’s Kappa coefficient (κ) [33], high inter-rater agreement coefficients were obtained (κ= 0.90). Cases of disagreement were discussed until a consensus was reached.
RESULTS
We first compared differences of autobiographical specificity as assessed with the TEMPau scale. We then compared differences on phenomenological characteristics between AD participants and older adults and within each population. Due to the scale nature of variables and their abnormal distribution, non-parametric tests were conducted. Between-groups comparisons were performed using Mann-Whitney’s U test and within groups comparisons were performed using Wilcoxon signed rank test. For all tests, the level of significance was set as p≤0.05, p values between 0.051 and 0.10, if any, were considered as trends. When reporting significant values, effect size was provided according to Cohen’s d criterion [34] (0.20 = small, 0.50 = medium, 0.80 = large).
Poor autobiographical specificity in AD participants
Analysis showed lower specificity in AD than in control participants (Z = – 4.93, p < 0.001, Cohen’s d = 1.55), with a Mean of 2.89 (SD = 0.87) and 3.90 (SD = 0.30), respectively.
Poor phenomenological reliving in AD participants (except for significance and emotion)
Table 2 depicts phenomenological characteristics of future thinking, as rated by each population. Compared to controls, AD participants reported lower reliving (Z = – 2.24, p < 0.05, Cohen’s d = 0.61), travel in time (Z = – 2.83, p < 0.01, Cohen’s d = 0.92), visual imagery (Z = – 5.39, p < 0.001, Cohen’s d = 1.82), auditory imagery (Z = – 2.45, p < 0.05, Cohen’s d = 0.69), language (Z = – 2.33, p < 0.05, Cohen’s d = 0.64), spatial specificity (Z = – 3.68, p < 0.001, Cohen’s d = 1.07), and temporal specificity (Z = – 3.35, p < 0.001, Cohen’s d = 0.97). No significant differences were detected for significance (Z = – 1.03, p > 0.1) or emotion (Z = – 0.37, p > 0.1) between both populations.
Phenomenological characteristics of autobiographical recall in Alzheimer’s disease (AD) and older adults
Note. Standard deviations are given between brackets; the maximum score was 4 points;
Poor visual imagery in AD participants
For each population, we carried out Friedman’s test to investigate differences in repeatedly measured phenomenological variables; this test showed significant differences for AD participants [χ2(8, N = 28) = 61.45, p < 0.001] and older adults [χ2(8, N = 31) = 25.73, p < 0.001].
As for AD participants, post-hoc signed rank tests by Wilcoxon showed significant differences between reliving and visual imagery (Z = – 3.89, p < 0.01, Cohen’s d = 1.28), reliving and language (Z = – 2.34, p < 0.05, Cohen’s d = 0.70), reliving and emotion (Z = – 2.77, p < 0.01, Cohen’s d = 0.69), travel in time and significance (Z = – 2.67, p < 0.01, Cohen’s d = 0.63), travel in time and visual imagery (Z = – 2.94, p < 0.01, Cohen’s d = 0.98), travel in time and emotion (Z = – 3.09, p < 0.01, Cohen’s d = 0.87), significance and visual imagery (Z = – 4.06, p < 0.001, Cohen’s d = 1.82), significance and auditory imagery (Z = – 3.45, p < 0.001, Cohen’s d = 0.97), significance and language (Z = – 3.13, p < 0.001, Cohen’s d = 1.21), significance and spatial specificity (Z = – 2.98, p < 0.01, Cohen’s d = 0.97), significance and temporal specificity (Z = – 3.04, p < 0.01, Cohen’s d = 0.96), visual imagery and auditory imagery (Z = – 2.27, p < 0.05, Cohen’s d = 0.71), visual imagery and language (Z = – 2.48, p < 0.05, Cohen’s d = 0.65), visual imagery and emotion (Z = – 4.31, p < 0.001, Cohen’s d = 2.13), visual imagery and spatial specificity (Z = – 2.72, p < 0.01, Cohen’s d = 0.81), visual imagery and temporal specificity (Z = – 2.89, p < 0.01, Cohen’s d = 0.85), auditory imagery and emotion (Z = – 3.53, p < 0.001), language and emotion (Z = – 3.98, p < 0.001, Cohen’s d = 1.33), emotion and spatial specificity (Z = – 3.91, p < 0.001, Cohen’s d = 1.24), emotion and temporal specificity (Z = – 3.98, p < 0.001, Cohen’s d = 1.40). All remaining comparisons were non-significant. Taken together, these analyses showed lower rating for visual imagery relative to remaining phenomenological features in AD participants. Conversely, these participants showed higher ratings for emotion and significance.
Regarding control participants, post-hoc comparisons showed significant differences between reliving and language (Z = – 2.56, p < 0.05, Cohen’s d = 0.71), travel in time and auditory imagery (Z = – 2.69, p < 0.01, Cohen’s d = 0.74), travel in time and language (Z = – 2.81, p < 0.01, Cohen’s d = 0.82), significance and auditory imagery (Z = – 3.22, p < 0.001, Cohen’s d = 0.84), significance and language (Z = – 3.21, p < 0.001, Cohen’s d = 0.90), visual imagery and auditory imagery (Z = – 2.81, p < 0.01, Cohen’s d = 0.57), visual imagery and language (Z = – 2.84, p < 0.01, Cohen’s d = 0.66), auditory imagery and emotion (Z = – 3.59, p < 0.01, Cohen’s d = 1.00), language and emotion (Z = – 3.34, p < 0.001, Cohen’s d = 1.05), language and spatial specificity (Z = – 2.28, p < 0.05, Cohen’s d = 0.62), emotion and temporal specificity (Z = – 2.49, p < 0.05, Cohen’s d = 0.65). All remaining comparisons were non-significant.
DISCUSSION
The main aim of this paper was to provide a fine-grained dissection of phenomenological characteristics of future thinking in AD. Relative to controls, AD participants showed poor rating for reliving, travel in time, visual imagery, auditory imagery, language, and spatiotemporal specificity, but no significant differences were observed between both populations for significance and emotion. Results also showed lower rating for visual imagery relative to remaining phenomenological features in AD participants; conversely, these participants showed higher ratings for emotion and significance.
Before we discuss our AD participants’ (subjective) phenomenological rating, it is important to highlight their (objective) autobiographical performance. Our analyses demonstrate lower autobiographical specificity in AD than in control participants, reflecting poor ability to construct detailed future scenarios in AD patients. These findings match research suggesting compromised ability in AD patients to project themselves into the future to contemplate hypothetical future scenarios [11–14]. Besides demonstrating a compromise in the ability to construct specific future scenarios, our findings demonstrate a compromise in several phenomenological dimensions of future thinking in AD.
With regards to reliving and travel in time, our findings demonstrate lower rating of these phenomenological characteristics in AD participants than in controls. Hence, AD seems to be associated with a weakened ability to mentally project oneself to relive future events (i.e., a decline in autonoetic consciousness). This assumption extends prior research suggesting a reduction of reliving for past thinking in AD. In the latter research, subjective reliving has been assessed with the “Remember/Know” paradigm, on which participants are typically asked to provide a “Remember” response if they were able to recover a specific event with its encoding context, or a “Know” response if they just knew this event had happened to them, but could not recall any contextual detail; AD patients tend to attribute less “Remember” than “know” responses for past events [35–37]. The compromise of subjective reliving in AD, as assessed with the “Remember/Know” paradigm, can be attributed to several factors such as the passage of time, the repetition of similar events, and the advancement of the disease [38]. These factors lead to substantial loss of episodic autobiographical details, and consequently, to a de-contextualization of autobiographical memories and a shift from the ability to mentally relive past events to a general sense of familiarity [38]. The latter suggestion is of interest as it links the decline of subjective reliving to compromise in autobiographical specificity. In our AD participants, the compromise of specificity was observed for (objective) analysis of specificity as well as for subjective rating of both spatial and temporal specificity.
Besides demonstrating lower rating than controls for reliving, travel in time, spatial specificity, and temporal specificity, our AD participants demonstrated lower rating for auditory imagery. According to William et al. [39], imagery in modalities other than visual (e.g., auditory) may be present but less common in autobiographical remembering. Poor auditory reliving may also be responsible for the low score for the statement “I or other people are talking” (i.e., language).
Among all the phenomenological characteristics, AD participants showed specific low rating for a core element of autobiographical reliving, which is visual imagery. It has been widely suggested that autobiographical memories are dominated by visual imagery and that the retrieval of visual images is the core feature of autobiographical reliving [40, 41]. In line with this idea, neuropsychological studies have also shown that lesions resulting in a loss of the ability to generate visual images also result in retrograde amnesia [42–44]. In a similar fashion, neuroimaging studies have shown activation of visual brain areas during autobiographical recall [45]. Moreover, there is evidence of low autobiographical recall in blind individuals [46]. These findings suggest visual processing is a core component of phenomenological reliving during autobiographical recall. As for AD, one study has found that whereas younger adults reported viewing their autobiographical memories as a continuous video, the AD group reported imagery resembling static snapshots akin to photographs or hazy imagery and lacking a real-life three-dimensional quality [47]. Interestingly, an impairment of phenomenological reliving for past thinking in AD was found to be especially pronounced for visual imagery [15], thus replicating the latter outcomes for future thinking. In our view, a compromise of visual imagery in AD results in decline in the ability to maintain and integrate visual images into a future scene.
Unlike their poor rating for reliving, travel in time, visual imagery, auditory imagery, language, and spatiotemporal specificity, our AD participants demonstrated a high rating for emotion and significance. In a broad manner, studies suggest better retention of emotional information in AD, especially if self-related. For instance, Sundstrøm [48] asked AD participants to retain neutral items and emotional self-related items (i.e., gifts to the participants). The author observed better recall for emotional items than for neutral items. In a similar fashion, Kalenzaga et al. [49] asked AD patients to rate neutral and emotional adjectives describing themselves, the authors observed better recall for emotional adjectives than for neutral adjective. Another study reported that AD patients can experience prolonged states of emotion that persist well beyond the patients’ memory for the events that originally caused the emotion [50]. As for future thinking, AD patients seem to construct future scenarios that are emotionally laden, which may allow them to regulate their affect and/or steer away from the present. The high emotional and significance values in future thinking in AD may also be a repetition of values in past scenarios. As the literature suggests, AD patients tend to replicate past scenarios for future thinking [11, 14].
As for control participants, research suggests that older adults tend to produce fewer specific details relative to young adults when imagining future events [51–53], an effect that may reflect age-related changes in episodic memory or executive functioning [54]. The phenomenological pre-experiencing of future events has also been found to be compromised in normal aging, which has been taken to reflect an increased reliance upon semantic representations compared to younger adults [51, 55]. In our study, and similar to AD participants, older adult controls showed high emotional rating for future thinking. This finding is in agreement with research suggesting high recall for emotional information in memory for past events in normal aging [56, 57]. It is noteworthy that the age difference in episodic memory performance can be, somehow, narrowed by emotion [58].
One limitation of our paper may lay in the fact that participants generated only one future event. Future replication should address this shortcoming by assessing phenomenological characteristics in a variety of future scenarios.
Future thinking compromise in dementia have potential implications for several complex functions that have crucial bearing on everyday living, functions such as prospective memory, decision-making, and maintaining a coherent sense of self across subjective time [10]. Besides these implications, future thinking seems to trigger a weakened phenomenological reliving in AD, except for significance and emotion. By highlighting the phenomenological experience as triggered by future thinking in AD, our study opens a window into the conscious experience of the future in these patients.
Footnotes
ACKNOWLEDGMENTS
Dr. El Haj was supported by the LABEX (excellence laboratory, program investment for the future) DISTALZ (Development of Innovative Strategies for a Transdisciplinary approach to Alzheimer disease). and the EU Interreg 2 Seas Programme 2014-2020 (co-funded by the European Regional Development Fund).
