Edited by: Charlotte A. Boettiger, University of North Carolina, Chapel Hill, USA
Reviewed by: Kelly Giovanello, University of North Carolina, Chapel Hill, USA; Francesca M. M. Citron, Lancaster University, UK
*Correspondence: G. Mioni, Department of General Psychology, University of Padova, Via Venezia, 8, 35131 Padova, Italy
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution and reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
The present study investigated the effect of Parkinson’s disease (PD) on prospective memory (PM) tasks by varying the emotional content of the PM actions. Twenty-one older adults with PD and 25 healthy older adults took part in the present study. Participants performed three virtual days in the Virtual Week task. On each virtual day, participants performed actions with positive, negative or neutral content. Immediately following each virtual day, participants completed a recognition task to assess their retrospective memory for the various PM tasks. PD patients were less accurate than the control group at both PM accuracy and recognition task accuracy. The effect of emotional valence was also evident, indicating that all participants were more accurate on positive PM tasks than both negative and neutral. This study confirmed PM impairment in PD patients and extended previous research showing how positive emotional stimuli can influence PM performance.
Prospective memory (PM) refers to memory for future intentions and involves remembering to perform an action in response to a specific cue, while being involved in an ongoing activity (McDaniel and Einstein,
For example, when the PM task included words Clark-Foos et al. (
Similar emotional enhancement in PM performance was also observed when the PM emotional stimuli included images. For example, Altgassen et al. (
Taken together, it seems clear that stimuli with emotional valence increased PM performance with a greater enhancement of stimuli with positive emotional valence (Kliegel and Jäger,
As far as we know, only two studies have been conducted to investigate emotional enhancement in clinical populations. Rendell et al. (
In recent years, an increasing interest in PM performance in Parkinson’s disease (PD) patients has been observed. PD patients frequently report cognitive decline in various executive function domains such as: inhibition, switching, and planning (Lewis et al.,
Considering that PM is a multiphase and complex construct that relies on integrity of executive functions (i.e., planning forming and executing the intention, monitoring for the appropriate moment to initiate the intended action, inhibit the ongoing activities and switch from ongoing activities to perform the intended action; McDaniel and Einstein,
For example, Kliegel et al. (
Considering the studies that used event- and time-based tasks, PD patients were less accurate than controls in both PM tasks (Ramanan and Kumar,
Methodological differences between these studies may explain the different results; but also the notion that particular features of PM tasks can influence PM performance has begun to guide more refined experimental studies (McDaniel and Einstein,
Interestingly, as previously introduced, Altgassen et al. (
Stimuli with emotional valence can be better and easily detected and can boost PM performance (Kliegel and Jäger,
So far, no study has explored the influence of emotional cues on PM performance in PD patients. To investigate the possible benefits of emotional cues on PM performance the content of the PM actions in the intention formation phase were manipulated. In particular, actions with positive, negative and neutral content were selected and assigned each action to an associated image with the same emotional valence (i.e., “Pay a speeding fine” was associated with an image showing a person given a speeding fine by a police officer). Following previous findings by Rendell et al. (
The present study had two main aims. Firstly, the investigation of PM performance in PD patients using a computerized task that simulated every day activities. Consistent with previous findings, PM dysfunctions in PD patients was predicted. In particular, the present study aimed to investigate if the PM dysfunction observed in PD patients is mainly caused by a dysfunction at the prospective or retrospective memory component of PM process. To this end, a recognition test of the PM content was included at end of each virtual day. According to the literature, we predicted that PM dysfunction in PD patients is mainly caused by a dysfunction at the PM component rather than at the retrospective memory component. Secondly, the effects of emotional cues on PM performance were investigated and the emotional content of the PM task was manipulated at encoding with expectation of an emotionally related improvement in PM performance. More precisely, we investigated whether PM actions with emotional valence will be better encoded and remembered compared to PM actions with neutral valence, in particular we predicted a positive enhancement. It was further tested whether the presentation of emotional images will interact with group (PD, controls), possibly attenuating PD-related PM dysfunctions.
Twenty-one older adults with PD (
PD group |
Control group |
|||
---|---|---|---|---|
Age (years) | 68.95 (5.38) | 71.12 (6.37) | 1.23 | 0.36 |
Education (years) | 7.76 (3.90) | 7.88 (3.05) | 0.11 | 0.03 |
MMSE | 27.28 (1.85) | 28.00 (.87) | 1.74 | 0.49 |
FAB | 15.95 (1.71) | 16.32 (1.14) | 0.87 | 0.25 |
BDI | 3.53 (3.69) | 2.92 (7.21) | 0.33 | 0.10 |
Years of disease duration | 5.7 (4.33) | |||
UPDRS (on stable medication) | 10.92 (5.55) | – | – | – |
NPI | 4.75 (6.75) | – | – | – |
The
The two groups did not differ significantly with respect to age, years of education, MMSE, FAB and BDI (
Participants were tested individually during two sessions that lasted approximately 90 min each: during the first session participants performed the neuropsychological tasks while during the second session participants performed the Virtual Week task. PD participants were tested while on their regular anti-PD medications. Demographic information for both groups was obtained through interviews and PD-related clinical characteristics were obtained from clinical chart reviews. Informed consent was collected from all participants and the study was conducted in accordance with Helsinki Declaration (59th WMA General Assembly, Seoul, 2008) and the guidelines of Department of General Psychology (Padova, Italy).
The present study used a computerized version of Virtual Week (Rendell and Henry,
Each event-based task was presented with a task relevant photo from the International Affective Picture System (IAPS; Lang et al.,
Immediately following each virtual day, participants completed a recognition test to assess their retrospective memory for the various PM tasks. The test required matching each intended action with its cue (i.e., PM action = “Pay a speeding fine”; PM cue = “shopping”). Participants were presented with a list of eight actions (e.g., pick up dry-cleaning), four of them were required during the virtual day and four were distracters. For each task, participants selected the matching cue from a pull down menu, listing the cues (e.g., when shopping, at university). The list of cues includes “not required” as one of the options to select, which is the correct response for the distractors that were included in the list of actions. Proportion correct was calculated for each emotional valence task (positive, negative and neutral). We considered false alarm when the participant recognized a distractor as a PM actions included in the virtual day.
Participants also performed a battery of neuropsychological tests that evaluate non-verbal intelligence, executive and memory functions with reference to normative data in the Italian population.
PM participants’ performance was analyzed in terms of proportion of correct responses. This was the number of correct responses, expressed as proportion of the four PM tasks scheduled for each of the three categories of emotional cues: positive, negative, and neutral. Data were analyzed with a 2 × 3 mixed ANOVA with the between-subjects variable of
To clarify the impact of emotional valence: positivity and negative enhancement/impairment indices were calculated as the difference between proportion correct on the positive minus the neutral tasks, and the negative minus the neutral tasks. This is consistent with Murphy and Isaacowitz (
Participants’ performance on recognition test was also analyzed in terms of proportion of correct responses of the four PM tasks in each category: positive, negative, and neutral (see also Terrett et al.,
Separate
Correlation analyses were also conducted separately for PD patients and controls to investigate the relationship between PM accuracy and accuracy at the recognition task and to investigate the involvement of cognitive abilities and clinical measures on PM performance and performance at the recognitiontask.
Participants’ performance as function of the emotional valence of the PM tasks and groups are reported in Figure
Group | Emotional valence | PM (accuracy) |
PM content recognised |
---|---|---|---|
PD patients | Positive | 0.78 (0.22) | 0.88 (0.17) |
Negative | 0.69 (0.30) | 0.75 (0.23) | |
Neutral | 0.64 (0.29) | 0.71 (0.23) | |
Control group | Positive | 0.93 (0.11) | 0.95 (0.10) |
Negative | 0.75 (0.20) | 0.89 (0.16) | |
Neutral | 0.84 (0.19) | 0.86 (0.22) |
The positivity enhancement was greater than zero for PD patients,
Participants’ performance on recognition test is reported in Figure
Analysis of responses to distractors showed that both groups were at ceiling with respect to accuracy on rejecting distractors: PD patients (
As with the accuracy on PM tasks, to clarify the impact of emotional valence on the retrospective memory test, positivity and negative enhancement/impairment indices were analyzed (see Figure
Table
PD patients |
Control group |
|||
---|---|---|---|---|
24.53 (5.69) | 27.45 (3.27) | 2.05* | 0.62 | |
36.22 (7.89) | 42 (6.79) | 2.57* | 0.78 | |
Part A | 84 (52) | 73 (28) | 0.93 | 0.27 |
Part B | 130 (56) | 147 (54) | 0.99 | 0.31 |
B–A | 53 (49) | 74 (53) | 1.29 | 0.41 |
Immediate recall | 33.50 (7.72) | 34.36 (8.57) | 0.35 | 0.10 |
Delayed recall | 6.85 (2.85) | 8.68 (2.46) | 2.31* | 0.68 |
Immediate recall | 13.37 (5.34) | 15.36 (4.58) | 1.33 | 0.40 |
Delayed recall | 14.74 (5.11) | 18.56 (4.77) | 2.55* | 0.77 |
Correlation analyses were also conducted separately on PD patients and controls between PM accuracy (regardless of valence) and accuracy at the recognition task (regardless of valence). Significant correlations were observed in both PD patients (
Correlations were also conducted with neuropsychological tasks. In the case of PD patients, significant correlations were observed between MMSE (
Moreover, correlations were also conducted, in PD patients, with clinical indices (NPI, UPDRS and disease duration). Significant correlation were observed between NPI
The purpose of the present study was to investigate PM performance in PD patients and healthy controls. Interestingly, the present study also examined the differences between groups in relation to the beneficial effects of emotional PM cues on PM performance. To this end, we used the Virtual Week task, a computer based task that simulates daily life activities and the effects of emotional cues on PM performance in individuals with PD and healthy participants were also compared. Finally, we investigated the retrospective memory component of PM by including a recognition task at the end of each virtual day.
Previous studies showed PM dysfunctions in PD patients in both time-based (Costa et al.,
Typically, on event-based PM tasks, participants are instructed to perform a specific action when presented with a cue that is embedded in an ongoing activity. Event-based PM tasks are generally considered less cognitively demanding than time-based PM tasks because event-based tasks are triggered by an external cue, whereas time-based tasks involve self-initiated processes to monitor the time (McDaniel and Einstein,
It is important to note that the event-based tasks used in the present study were different every virtual day and simulated the type of
PD patients often present executive dysfunction, even in the early stages of the disease (McKinlay et al.,
Although some studies have reported that retrospective problems do not interfere with PM performance in PD, very simple paradigms were used in those studies in which participants were asked to recall the instructions of the PM task (Katai et al.,
Further support for cognitive impairment being one of the possible causes of PM dysfunction in PD patients, comes from the significant correlations observed between measures of executive functions and PM accuracy and accuracy during the recognition task. In particular, in the case of PM accuracy, significant correlations were observed with indices of cognitive efficiency (MMSE) and executive functions (Semantic fluency); whereas in the case of the recognition task, significant correlation was observed with an index of retrospective memory (delayed recall in the prose recall task). The results confirmed previous findings highlighting the involvement of executive functions in PM performance (Kliegel et al.,
The correlations of clinical indices are also of great interest in relation to the study of PM performance in PD patients. Our study showed that patients with more severe PD were also less accurate on both PM accuracy and accuracy at the recognition task.
The present study also addressed the question of emotionally related improvement in PM performance and this is the first study that investigated this issue in PD patients. We predicted that PM actions with emotional valence would be better performed compared to PM actions with neutral valence; in particular we expected a greater enhancement of stimuli with positive emotional valence (Kliegel and Jäger,
We also calculated positivity and negativity effects (Murphy and Isaacowitz,
Previous studies conducted with healthy older adults consistently showed the emotional effects on PM performance but the direction of these effects were diverse. Altgassen et al. (
It is possible that in forming and carrying out PM intentions, positive activities receive higher priority than negative intentions. This hypothesis fits our everyday experience; pleasant activities are more likely to be executed compared to unpleasant activities that are postponed and/or “forgotten”. More specific interpretation can be given considering the specificity of PD population.
A large number of different structures are involved in recognizing emotional stimuli: the occipitotemporal cortices, amygdala, orbitofrontal cortex, basal ganglia, and right parietal cortices, among others, in particular the dopaminergic system is believed to be critical in the emotion recognition process (Adolphs,
A limitation in the present study is the difference in the level of education in younger and older participants. However, it might be noted that this discrepancy is representative of the Italian population. Italian older adults often present with lower education levels as the majority of older adults only completed primary school (Mioni et al.,
In summary, our study showed that PD patients were less accurate than control group in PM tasks. Also PD patients were less accurate than control group at recognition tasks indicating that PM dysfunction observed in PD patients is partially caused by retrospective memory dysfunction. The correlation analyses also showed that executive dysfunctions are involved in PM performance, indicating that PD patients with lower cognitive resources were also less accurate. These results were also confirmed by the correlations observed with the indices of severity (UPDRS and NPI). Interestingly, our results also showed positive enhancement of PM performance on event-based PM tasks. The positive enhancement observed in PD patients can have interesting implications for rehabilitation programs. In fact, future studies should further investigate the effects of positive stimuli in PD patients in PM performance and manipulate the presentation of the emotional cue at the encoding or retrieval phase.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
1Focal/non-focal PM cues refer to the link between PM cue and ongoing task. If performing the ongoing task requires focal processing of the PM cue, then the cue is sufficiently processed to enable involuntary (automatic) retrieval of the intended action. In other cases, if the ongoing task does not direct attention toward processing the features of the PM cue than the PM cue can be considered nonfocal.
2Lower score at NPI represents better performance.