Sage Journals: Discover world-class research

Abstract

Prospective memory (PM), the ability to plan and execute intentions in the future, plays a critical role in managing everyday tasks. A gap exists in our understanding of the neural mechanisms underlying PM retrieval based on semantic judgment, particularly when using picture-based stimuli. The current study tested 2 novel picture-based semantic-judgment PM tasks: Animal-cued Prospective Retrieval Task (Ac-PRT) and Object-cued Prospective Retrieval Task (Oc-PRT), designed to investigate PM processes involved in intention formation (Cue trials), intention retention (Ongoing trials), and intention retrieval (PM Retrieval trials). Twenty-three young adults, aged 18 to 30 years, completed the tasks while EEG data were recorded. Behavioral results showed that participants responded more slowly during Ongoing trials compared to Cue and PM Retrieval trials and were less accurate during PM Retrieval trials. Additionally, between the tasks, responses were faster and more accurate in Ac-PRT than in Oc-PRT during both Ongoing and PM Retrieval trials. ERP analyses revealed distinct neural signatures across trial types, particularly in P2, N2, N4, and Parietal Positivity (PP) components in both tasks. Additionally, task-specific differences were observed during the PM Retrieval trials in P2, N4, and PP amplitudes and in PP amplitude during the Ongoing trials. These findings demonstrate that the 2 tasks effectively dissociated core PM processes and showed category-specific differences in behavioral and neural mechanisms, offering a robust framework for future investigations of PM in aging and clinical populations.

Keywords

cognition electrophysiology prospective memory future intentions young adults ERP semantic judgment event-related potentials

Introduction

Prospective memory (PM) refers to one’s ability to plan an action and perform it in the future.¹ Remembering to go to an appointment, taking medication, or picking up dry cleaning on the way back from work are examples of everyday tasks that are supported by PM. Prospective memory retrieval is a complex, multistage process that involves: (i) forming an intention, (ii) retaining the intention over an interval, (iii) initiating and executing the intended action, and (iv) evaluating the outcome.² For instance, after forming the intention to pick up dry cleaning after work, one must retain both the intention and the plan while continuing with daily activities, consider the best route based on evening traffic, execute the task after work, and ensure the correct piece of dry cleaning was picked up. The timing and manner of task execution depend on the cue that triggers it. PM retrieval is typically classified as either event-based or time-based, depending on the type of cue.³ While event-based retrieval is triggered by external events, such as the end of one’s workday in the previous example, time-based retrieval is driven by internal cues linked to a specific time, such as picking up dry cleaning at 6 pm.

Theoretical postulations have explored PM through various lenses. While traditional models have viewed PM as a discrete event, either time- or event-based,¹ newer models suggest that it is more context-driven and dynamic.^4
-6 The Multiprocess model suggests that PM retrieval can involve automatic processes for simpler tasks, or strategic processes for more complex tasks, and depends on cues.⁴ The Preparatory Attention and Memory Model (PAM) suggests that PM retrieval involves 2 distinct stages: (i) preparatory attention stage, which involves focusing on the ongoing task while preparing for the upcoming cue, and (ii) cue-triggered retrieval stage that occurs when the cue is presented, and attention is maintained during the retention period to enhance retrieval success.⁵ The Dynamic Process Model emphasizes that PM retrieval is a flexible, continuous process where cognitive control and attention are dynamically adjusted based on task complexity and changing cues.⁶ Finally, the Attention to Delayed Intention (AtoDI) model explains the brain mechanisms underlying the processes outlined in the PAM and Multiprocess models.⁷ The model highlights the dissociation between intention maintenance and retrieval phases in the context of top-down and bottom-up attentional processes. Overall, theoretical postulations and their supporting evidence suggest that PM retrieval engages several cognitive functions, including attention,^8
-10 episodic memory, and cognitive control.^10
-14 Attention is critical for forming and sustaining intention driven by internal and/or external goals.⁷ It helps one to remain focused on cues that suggest it is time to act on the intentions.¹⁵ Episodic memory, on the other hand, is essential for recalling the specific details of the intended action¹⁶ (eg, what, when, and where). Cognitive control processes including inhibition, working memory, cognitive flexibility, and goal maintenance are essential for planning, prioritizing, and adapting behaviors to align with one’s goal.^10
-14 Collectively, these cognitive functions work together, supporting retrieval and execution of intentions within the broader context of the ongoing task.

Growing neuroimaging research has unraveled the neural underpinnings of PM, emphasizing the role of frontal regions.^7,17
-23 Activation of dorsal and ventral frontoparietal networks has been observed during successful PM retrieval, along with increased activation in the left rostral prefrontal cortex during time-based compared to event-based tasks.^7,24 The rostral prefrontal cortex has been shown to play a role in directing attention to the stimulus, while medial and lateral prefrontal regions have been found to be important for balancing attention between ongoing stimuli and PM intentions.^17,18,25
-29 More recently, evidence suggests that distinct neural networks are differentially engaged depending on whether attention or memory demands dominate a PM task.³⁰ While these neuroimaging studies have advanced our understanding of the neural substrates linked to PM, capturing the real-time dynamics of cognitive processes involved in PM remains a challenge. Electroencephalography (EEG), an affordable and widely accessible technique with high temporal resolution, provides valuable insights into the timing of cognitive processes and underlying neural mechanisms linked to PM.

West and Covell were the first group to explore PM in younger and older adults using an EEG paradigm consisting of ongoing, lure, and retrieval trials.³¹ Ongoing trials involved word pairs written in lowercase letters, and participants were asked to judge whether the words were related to each other by pushing a button. PM retrieval trials consisted of uppercase word pairs, requiring participants to detect the cue (letter case) and press a different button than the relatedness judgment. The lure trials involved a mixed case word pair, requiring a relatedness judgment similar to ongoing trials, without needing to act on any stored intentions. Over the years, subsequent research has progressively deepened our understanding of the associated processes. Several studies have examined aspects of PM such as the encoding of intention,^2,7,32,33 intention load,³⁴ intention retention by varying the time between formation of an intention and its execution³⁵ or by varying the storage demands that is, cognitive effort required to keep the intention in mind over time.^7,36
-42 Additionally, studies have examined the effect of focality of cue^5,6,43
-45and perceptual or conceptual distinctiveness of cue^36,46
-50 on PM retrievals. The focality of a PM cue has been shown to significantly influence the attentional resources required for successful retrieval.⁵¹ In focal PM tasks, both the cue and the ongoing task engage the same cognitive processes. This overlap allows the intended action to be retrieved more automatically, with minimal attentional demands. In contrast, non-focal PM tasks involve cues that engage cognitive processes different from ongoing task, requiring individuals to engage in more effortful, strategic monitoring to detect the cue and retrieve the intention.⁴ However, recent perspectives suggest that the distinction between focal and non-focal PM tasks may not be strictly categorical. Instead, tasks can be conceptualized along a continuum, with varying degrees of overlap between cue and ongoing task processing. This continuum view reflects the nature of most everyday PM tasks, which often involve a combination of bottom-up (stimulus-driven) and top-down (goal-directed) processes. Rather than relying exclusively on one mechanism, individuals may flexibly engage both depending on task demands, cue salience, and contextual factors.^7,45

EEG studies have examined PM in the context of target detection, N-back working memory, continuous recognition, and semantic judgment.^39,50,52
-55 Most of these studies have manipulated perceptual features of the stimuli, including letter case, color, letter identity, and word identity.^7,42,48 Some have manipulated semantic categories of words.^48
-50,56 EEG PM studies^{31,34,36,47,50,53,55,57
-59} have identified several ERP components (see Table 1), which vary based on the type of task and stimuli. For instance, perceptual manipulations (eg, letter case) have been associated with N300 and parietal positivity in young adults, while conceptual manipulation (eg, semantic judgment) has yielded a delayed⁵⁰ or absent⁵⁶ N300. Additionally, different versions of ongoing and PM retrieval tasks have yielded distinct findings for Late Parietal Positivity Complex (LPC).⁶⁰ For example, a task-switching version (participants switched between different tasks) yielded a late positivity, whereas a dual-task version (participants performed 2 tasks simultaneously) failed to yield this late positivity.

Table 1.

ERP Components Linked to PM Retrieval and Associated Time Windows.

ERP component	Time window (ms)	References
P2	150-275	Brewer et al, 2010⁴⁶; Rumsey et al, 2022⁵⁵
N2	200-300	Hockey & Cutmore, 2021¹⁰⁰
N300	300-500	West et al, 2001⁵²; Wang et al, 2013⁴⁸
N400	350-550	Cona et al, 2015⁷; Cruz et al, 2016⁵⁰
Frontal positivity	300-500	West et al, 2001⁵²; Cejudo et al, 2022⁵⁹
Parietal/prospective positivity/late parietal positivity complex	400-800	West and Covell, 2001⁵²; Hockey & Cutmore, 2021¹⁰⁰; Cejudo et al, 2022⁵⁹
Fronto-polar slow waves	500-1000	West et al 2003²; Hering et al, 2020³⁴
Temporal-parietal slow waves	800-1200	West et al, 2003²; Hering et al, 2020³⁴
RON (Reorientation Negativity)	400-750	Rumsey et al, 2022⁵⁵

Despite this growing body of work, EEG studies examining PM have not always considered the impact of certain factors such as (i) reading ability (since the classic PM paradigms require participants to read letters/words/word pairs), (ii) frequency of incongruent stimuli that make up the Lure and PM Retrieval trials, (iii) number of stimuli between intention-formation trials and PM retrieval trials, and (iv) presence/absence of Lure trials. Also, only a few PM EEG studies have used semantic categorization judgment (eg, animal vs object)^39,50,52
-55 tasks and none have used picture stimuli while controlling for the variables mentioned above.

To address these gaps, we developed 2 novel PM tasks involving basic categorization judgments related to animals and objects using decontextualized line drawings, while carefully accounting for key design variables such as stimulus congruency, trial structure, and inclusion of lures. The primary aim of this study was to evaluate the feasibility of disentangling neural mechanisms underlying intention formation and PM retrieval, indexed by ERPs, for each of the 2 novel tasks. To achieve this, we incorporated different trial types: PM Retrieval (intention retrieval and execution phase), Ongoing (retention phase), Cue (intention encoding phase), and Lure trials (distractor during retention phase), within each task, to assess various aspects of intention formation and PM retrieval. This task falls between focal and non-focal PM paradigms, aiming to improve its ecological validity by better reflecting everyday PM tasks. We anticipated significant behavioral differences in accuracy and response times between different trial types, with slower or more error-prone responses in PM trials relative to all the other trial types. Additionally, we expected to observe neural differences in ERP components on measures of latency and amplitude.

Given substantial evidence that animal and object categories engage partially distinct cognitive and neural systems,^61
-64 a secondary objective of the study was to investigate the extent to which the semantic characteristics of the ongoing task modulate PM retrieval, that is, to examine task effects. We anticipated differences in behavioral measures of reaction time and accuracy across trials, and neural differences in ERP components on measures of latency and amplitude between the 2 tasks.

Method

Participants

Twenty-three young adults (18 F; 20.5 ± 2.35 years, mean education = 14.17 ± 1.89 years) recruited from the University of Illinois campus and Champaign-Urbana community participated in the study (see Table 2). All participants were native speakers of English, right-handed, and had a minimum of high school education. They had normal global cognitive functioning on the Montreal Cognitive Assessment⁶⁵ (a score of 27 or higher), with no history of learning disabilities, neurological illnesses, head trauma, major psychiatric illnesses, alcohol or substance abuse, uncorrected hearing loss, or vision loss. Written informed consent was obtained from all participants in accordance with Institutional Review Board protocol approved by the University of Illinois Urbana-Champaign, and data collection was performed in accordance with the ethical standards of this institution and the Helsinki Declaration of 1975.

Table 2.

Participant Demographics.

Participants (N = 23)	Mean (Standard Deviation)
Age (years)	20.5 (2.4)
Education (years)	14.2 (1.9)

Experimental Task

All participants completed two PM tasks that were developed in-house, Animal cued-Prospective Retrieval Task (Ac-PRT) and Object cued-Prospective Retrieval Task (Oc-PRT), during which EEG data were collected. Stimuli for both tasks consisted of black and white, line-drawn pictures of (i) animals and objects enclosed in 1 of 3 shapes: a square, a circle, or a triangle, and (ii) empty shapes: a triangle or a circle. All line drawings were selected and adapted from Snodgrass and Vanderwart⁶⁶ and the CRL-UCSD International Picture Naming Project (IPNP) Database,⁶⁷ which are large international repositories of standard picture stimuli. Both tasks included 90 line-drawings of animals and 90 line-drawings of objects, which were controlled for frequency (animals: M = 1.78, SD = 1.04; objects: M = 2.27, SD = 0.90) and syllable length (animals: M = 1.86, SD = 0.86; objects: M = 1.73, SD = 0.77). Both typical animals (eg, cat, cow, dog) and atypical animals (eg, whale, snake, worm) were included. Objects included clothing (eg, jacket, boot, scarf), household objects (eg, book, table, bowl), food (eg, cake, egg, cherry), musical instruments (eg, bell, drum, guitar), tools (eg, hammer, scissors, wrench), and vehicles (eg, bus, car, wheelchair).

All line drawings were resized to fit within 3 in × 3 in dimensions and were pilot tested to ensure accurate recognition. Participants responded using a 3-button response box (YES, NO, OTHER), which allowed for precise and consistent recording of response times while minimizing motor variability across trials. To account for individual differences in motor speed, all participants completed a baseline reaction time task involving the same response box. The average response times from this baseline were used to adjust final behavioral performance metrics.

Both tasks consisted of 4 types of trials: Ongoing Task Trials (referred to henceforth as Ongoing trials), Cue, Lure and PM Retrieval trials (see Figure 1).

Figure 1.

Exemplars of trial types.

Ac-PRT: This task included 229 stimuli (151 Ongoing, 30 Cue, 18 Lure, and 30 PM Retrieval trials). The Ongoing trials consisted of both animal (n = 30) and object (n = 121) stimuli, each enclosed in a square. Participants were instructed to judge whether each stimulus represented an animal by pressing the “YES” button for animals and the “NO” button for objects. The Cue trials, Lure trials, and PM Retrievals trials were interspersed among the Ongoing trials. The Cue trials consisted of empty shapes (circle/triangle). Participants were instructed to push the “OTHER” button when they saw an empty shape, signaling the formation of a PM intention. Participants were also instructed to continue pressing the “YES” button for animals and “NO” button for objects after seeing the empty shape. However, when an animal enclosed in the previously empty shape appeared, they were told to press the “OTHER” button again. Importantly, these trials in which an animal drawing appeared inside the previously empty shape (Cue trial), constituted the PM Retrieval trials. The Lure trials, on the other hand, consisted of pictures of animals within a shape different from the previous cue (eg, a cat in a circle when the cue shape was a triangle) or an object within the previously cued shape (eg, a shirt in a triangle). Lure trials were designed to share properties with the PM Retrieval trials, either in terms of the semantic category or shape. Participants were expected to respond to these trials in a manner similar to the Ongoing trials (see Figure 2 for sample sequence).

Figure 2.

Sample sequence in Ac-PRT.

Oc-PRT: The task included 229 stimuli (151 Ongoing, 30 Cue, 18 Lure, and 30 PM Retrieval trials). The Ongoing trials consisted of both objects (n = 30) and animals (n = 121), each enclosed in a square. Participants were instructed to judge whether each stimulus represented an object by pressing the “YES” button for objects and the “NO” button for animals. Like the Ac-PRT, the Cue trials, Lure trials, and PM Retrievals were interspersed among the Ongoing trials. The Cue trials consisted of empty shapes (circle/triangle). Participants were instructed to push the “OTHER” button for the Cue trials signaling the formation of a PM intention. Participants were also instructed to continue pressing the “YES” button for objects and “NO” button for animals after seeing the empty shape. However, when an object enclosed in the previously empty shape appeared, they were told to press the “OTHER” button again—these trials constituted the PM Retrieval trials. The Lure trials consisted of pictures of objects within a shape different from the previous cue (eg, a shirt in a circle when the cue shape was a triangle) or an animal within the previously cued shape (eg, a cat in a triangle). Participants were expected to respond to these Lure trials in a manner similar to the Ongoing trials (see Figure 3 for sample sequence).

Figure 3.

Sample sequence in Oc-PRT.

Four versions were created for both tasks, with each version being matched for frequency and syllable length of stimuli. These versions were assigned to participants in a pre-determined order to maintain balance across versions. For both tasks, the sequence of trials was pseudorandomized, such that each task began with Ongoing trials (2-3 trials), followed by a Cue trial, followed by Ongoing trials (0-5) that also included a Lure trial, and finally a PM Retrieval trial (see Figure 4). Sixty such trial sequences were used for each task. Each stimulus was presented for 1200 ms, with a 1200 ms interstimulus interval.

Figure 4.

Example trials sequence.

For both tasks, participants were given verbal and written instructions, which were then again shown on the computer screen before each task. As an example, the following instructions were given for the Ac-PRT task:

“In this task, you will see pictures of animals and objects. When you see a picture of an animal enclosed in a square, press the ‘YES’ button, if you see an object press the ‘NO’ button. If you see an empty shape, press the ‘OTHER’ button. Continue pressing the ‘YES’ button for animals and ‘NO’ button for objects after seeing the empty shape. However, if you see an animal enclosed in the shape which was previously empty, press the ‘OTHER’ button again.”

All participants completed a practice session with a criterion of 75% performance to ensure understanding of the task before completing both the tasks during which EEG data were recorded. Reaction time and accuracy of judgments were recorded, and the order of the tasks were counterbalanced across participants.

EEG Data Acquisition and Preprocessing

Continuous EEG data were recorded from a 64-electrode Neuroscan Quikcap using a Neuroscan SynAmpsRT amplifier and Curry 7 software (sampling rate: 1 kHz, DC-200 Hz), with electrode impedances typically below 10 kΩ. The reference electrode was located at the vertex between Cz and CPz. Vertical electrooculogram (VEOG) was recorded at sites above and below the left eye. Data were processed off-line using ERPLAB plug-in on EEGLAB on MATLABR2023a platform.^68,69 The continuous EEG data were high-pass filtered at 0.15 Hz and corrected for eye blinks using spatial filtering. The data were then epoched between 500 ms before stimulus onset to 1500 ms after stimulus onset. Epochs with peak signal amplitudes of ±75μ V were rejected. The EEG data were then re-referenced to the average potential over the entire scalp. Baseline correction was done by subtracting the mean amplitude of the pre-stimulus interval (–200 to 0 ms) from each time point. ERP data for each participant were averaged for the 3 trial types (Ongoing trials, which included animal trials for Ac-PRT and object trials for Oc-PRT, Cue trials, and PM Retrieval trials) separately for both tasks. Table 3 illustrates distribution of trials in both tasks along with the subset of trials considered for final analysis. Specifically, an equal number of epochs were analyzed from Ongoing trials (n = 30), Cue trials (n = 30), and PM Retrieval (n = 30) trials. Only trials with correct responses were included in the ERP averages, and trials with reaction times shorter than 200 ms or longer than 1000 ms were excluded from further analysis.

Table 3.

Trial Breakdown for Ac-PRT and Oc-PRT.

Trial type	Number of trials		Analysis
Ongoing	Ac-PRT	Oc-PRT	Ac-PRT	Oc-PRT
Ongoing	Animal: 30Object: 121	Animal: 121Object: 30	Animal: 30	Object: 30
Cue	30		30	30
PM Retrieval	30		30	30

ERP Analysis

Informed by the ERP literature on PM,^54,59,70 as well as visual inspection of the grand average ERPs, ERP components, specifically P2, N2, N4 at anterofrontal (AF3, AF4) and frontal (F1, Fz, F2) electrode clusters, and parietal positivity (PP) at central (C1, Cz, C2) and parietal (P1, Pz, P2) electrode clusters were examined. Electrode sites and time windows were estimated based on (i) previous studies reporting similar components^54,59,71,72 and (ii) the consistency with which the ERP components were observed across all participants on visual inspection. The following time windows were used for both Ac-PRT and Oc-PRT to identify the ERP components: (i) P2, between 75 and 250 ms, (ii) N2, between 175 and 350 ms (iii) N4, between 350 and 550 ms, and (iv) PP, between 350 and 650 ms. Peak latencies and mean amplitudes were examined.

Statistical Analysis

Statistical analyses were conducted in R 4.3.0 (R Core Team, 2023⁷³) for the Ac-PRT and Oc-PRT tasks. For reaction time (RT), linear mixed-effects models (lmer function, lme4 package⁷⁴) were used to examine the effect of trial on log-transformed reaction times (rt_log), with trial as a fixed effect and random intercepts for participants. Analysis of Variance (ANOVA) was conducted to assess the significance of trial (formula: RT ~ Trial + [1|Participant]). Post hoc pairwise comparisons of estimated marginal means were performed using the emmeans package, with degrees of freedom calculated via the Satterthwaite method. For accuracy (correct/incorrect), generalized linear mixed-effects models (GLMMs) were fitted using the glmer() function (lme4 package), with a binomial family (formula: Accuracy ~ Trial + [1|Participant]). Pairwise comparisons of estimated marginal means were performed, with P-values adjusted using the Satterthwaite approximation. Post hoc contrasts for accuracy were performed on the log odds ratio scale and tested using a Z-test, with no degrees of freedom reported.

For ERP data, a linear mixed-effects model was fitted to examine the effects of trial and electrode cluster, as well as their interaction for latency and amplitude measures, accounting for random participant-level intercepts. The model was estimated using maximum likelihood (ML) with the formula: [ERP measure ~ Trial * Electrode + (1|Participant)]. An ANOVA was conducted to assess the significance of the fixed effects, followed by pairwise comparisons of trial levels within each electrode cluster with Bonferroni adjustment for multiple comparisons.

To explore task-specific effects, behavioral and ERP responses across the two PM tasks (Animal-cued vs Object-cued) were compared. For RT, 1-way ANOVAs were performed on log-transformed RT values (rt_log) to assess the effect of trial, with Task as a between-subjects factor. Separate ANOVAs were conducted for each trial type (PM Retrieval, Ongoing, and Cue), allowing for comparisons of RT across task conditions within each trial category. Log transformation was applied to RT data to normalize distributions and meet assumptions of parametric testing. Accuracy data were analyzed using 1-way ANOVAs with Task as a between-subjects factor for each trial type. All ANOVA models were fitted using the anova_test() function from the rstatix package, and results included F-values, degrees of freedom, and P-values.

To examine neural dynamics associated with task performance, linear mixed-effects models were fitted to ERP measures recorded during Ongoing and PM Retrieval trials. Separate models were constructed for each ERP component, N2, P2, N4, and LPC, evaluating both peak latency and mean amplitude. Each model included Task as a fixed factor and Electrode Cluster as a covariate to account for spatial variability across scalp sites. Random intercepts for participants were included to account for subject-level variability. Models were estimated using maximum likelihood (ML) with the formula: [ERP measure ~ Task + Electrode + (1|Participant)]. ANOVAs were conducted to assess the significance of the fixed effects. Post hoc pairwise comparisons of task conditions were performed using the emmeans package, with P-values adjusted via Bonferroni correction to control for multiple comparisons.

Results

Behavioral Data: Trial Type Effects Within Tasks

Reaction Time

Linear mixed-model analysis of log-transformed RT showed significant differences across trial type for both Ac-PRT (F [5, 11200) = 398.95, P < .001] and Oc-PRT [F (5, 10899) = 634.76, P < .001]. Post hoc comparisons showed significantly slower reaction time for Ongoing compared to PM Retrieval trials [Ac-PRT (P < .001), Oc-PRT (P < .001)]. Additionally, faster RTs were noted for Cue compared to PM Retrieval trials [Ac-PRT (P < .001), Oc-PRT (P < .001)] and Cue compared to Ongoing trials [Ac-PRT (P < .001), Oc-PRT (P < .001)] (see Table 4 for mean RT and accuracy across trial types for Ac-PRT and Oc-PRT Behavioral Data).

Table 4.

Mean Reaction Times and Accuracy Across Trial Types for Ac-PRT and Oc-PRT.

Task	Trial	Raw RT mean (SD) (ms)	Log RT mean (SD)	Accuracy (%)
Ac-PRT	Cue	568.80 (155.32)	6.31 (0.25)	0.94 (0.22)
	Ongoing	656.30 (199.86)	6.44 (0.28)	0.91 (0.28)
	PM Retrieval	630.79 (193.93)	6.40 (0.28)	0.87 (0.33)
	Lure	736.23 (224.33)	6.56 (0.29)	0.89 (0.31)
Oc-PRT	Cue	569.57 (154.99)	6.31 (0.24)	0.96 (0.19)
	Ongoing	739.77 (197.48)	6.57 (0.25)	0.87 (0.33)
	PM Retrieval	678.09 (206.44)	6.47 (0.28)	0.84 (0.36)
	Lure	796.96 (211.34)	6.65 (0.25)	0.80 (0.40)

Abbreviations: RT, reaction time; SD, standard deviation.

Accuracy

For accuracy, significant differences among trials were observed for both Ac-PRT (z = −3.79, P < .001) and Oc-PRT (z = −9.63, P < .001). Post hoc comparisons showed that in Ac-PRT, accuracy of performance was higher for Ongoing compared to PM Retrieval trials (z = 3.813, P = .0019). Also, on both tasks, participants had higher accuracy for Cue compared to PM Retrieval trials (Ac-PRT [P < .001], Oc-PRT [P < .001]) and Cue compared to Ongoing trials (Ac-PRT [P < .001], Oc-PRT [P < .001]; see Table 4 mean RT and accuracy across trial types for Ac-PRT and Oc-PRT Behavioral Data).

Behavioral Data: Trial Type Effects Between Tasks

Reaction Time

Linear mixed-model analysis of log-transformed RT revealed significant main effects of Trial (F [1, 5942.7] = 106.10, P < .001), Task (F [1, 5942.3] = 218.51, P < .001), and a significant Trial × Task interaction (F [1, 5942.2] = 17.18, P < .001). Post hoc comparisons revealed significantly faster RT for both PM Retrieval and Ongoing trials in Ac-PRT compared to Oc-PRT (P < .0001). No significant difference was observed in reaction time for Cue trials across the 2 tasks (P = 1.000).

Accuracy

Significant main effects of Trial (F [1, 6782] = 18.31, P <.001]) and Task (F [1, 6782] = 24.73, P < .001) for accuracy were observed. Trial × Task interaction (F [1, 6782] = 0.44, P = .506) was not significant. Post hoc comparisons revealed higher accuracy for both PM Retrieval (P = .018) and Ongoing trials (P = .002) in Ac-PRT compared to Oc-PRT. No significant difference was observed in accuracy for cue trials across the 2 tasks (P = 1.000). Task Variability in behavioral performance across Ac-PRT and Oc-PRT, with focus on PM Retrieval trials, is illustrated in Figure 5. As reaction time increases in Ac-PRT, it also increases in Oc-PRT. Similarly, higher accuracy in Ac-PRT is associated with higher accuracy in Oc-PRT. This suggests that PM Retrieval processes are similarly engaged across both task formats.

Figure 5.

Task variability in reaction time and accuracy of PM Retrieval trials. (Reaction time in milliseconds; accuracy in %).

ERP Data: Trial Type Effects Within Tasks

Peak latency and mean amplitude measures across trial types for the ERP components (P2, N2, N4, and PP) for both Ac-PRT and Oc-PRT are reported in Table 5. Results related to ERP differences across trial types for latency and amplitude measures are reported in Tables 6 and 7, respectively. Grand average ERPs and scalp maps are illustrated in Figures 6 and 7.

Table 5.

Peak Latencies and Mean Amplitudes for the ERPs (P2, N2, N4, and PP).

ERP by Trial	Ac-PRT		Oc-PRT
	Peak latency (ms)	Mean amplitude (µV)	Peak latency (ms)	Mean amplitude (µV)
P2
Cue	152.17 (22.92)	0.52 (0.53)	157.77 (31.83)	0.59 (0.64)
Ongoing	145.14 (28.62)	−0.13 (0.52)	137.68 (39.03)	−0.09 (0.62)
PM Retrieval	144.41 (38.68)	0.47 (0.73)	149.78 (34.70)	0.15 (0.58)
N2
Cue	281.46 (24.11)	−0.16 (0.55)	286.36 (23.25)	−0.44 (0.84)
Ongoing	253.63 (34.10)	−0.66 (0.67)	254.55 (44.57)	−0.64 (0.96)
PM Retrieval	269.98 (35.03)	−0.48 (0.70)	270.39 (36.18)	−0.59 (0.75)
N4
Cue	441.50 (42.50)	−0.58 (0.44)	466.68 (48.13)	−0.84 (0.60)
Ongoing	436.43 (45.61)	−0.26 (0.40)	441.96 (46.26)	−0.33 (0.56)
PM Retrieval	453.88 (48.02)	−0.87 (0.61)	449.39 (46.68)	−0.65 (0.60)
PP
Cue	478.00 (65.56)	0.49 (0.36)	476.16 (64.34)	0.49 (0.43)
Ongoing	478.20 (59.29)	0.44 (0.36)	497.46 (74.21)	0.31 (0.30)
PM Retrieval	498.30 (74.34)	0.38 (0.32)	491.41 (68.41)	0.28 (0.32)

Cells represent mean (standard deviation).

Table 6.

Statistical Results for Peak Latency (P2, N2, N4, and PP).

ERP components	Ac-PRT			Oc-PRT
	F	P	η²p	F	P	η²p
P2
Trial	3.16	.04*	0.126	7.55	<.001*	0.255
Electrode	1.72	.19	0.073	1.47	.23	0.063
Interaction	0.42	.66	0.019	0.33	.72	0.015
N2
Trial	12.47	<.001*	0.362	14.19	<.001*	0.392
Electrode	3.62	.06	0.141	0.50	.48	0.022
Interaction	0.76	.47	0.033	1.40	.25	0.060
N4
Trial	2.52	.09	0.103	5.39	.01*	0.197
Electrode	7.32	.01*	0.250	6.46	.01*	0.227
Interaction	0.36	.70	0.016	1.82	.17	0.076
PP
Trial	2.65	.08	0.108	1.67	.19	0.071
Electrode	36.62	<.001*	0.625	18.24	<.001*	0.453
Interaction	0.34	.71	0.015	0.40	.67	0.018

Statistically significant.

Table 7.

Statistical Results for Mean Amplitude (P2, N2, N4, and PP).

ERP Components	Ac-PRT			Oc-PRT
	F	P	η²p	F	P	η²p
P2
Trial	34.36	<.001*	0.610	33.94	<.001*	0.607
Electrode	2.04	.16	0.073	4.63	.03*	0.063
Interaction	0.54	.58	0.019	0.70	.50	0.015
N2
Trial	22.93	<.001*	0.510	3.96	.02*	0.153
Electrode	32.69	<.001*	0.598	2.01	.16	0.153
Interaction	4.28	.02*	0.163	1.87	.16	0.060
N4
Trial	36.89	<.001*	0.626	30.84	<.001*	0.584
Electrode	0.19	.66	0.250	14.79	<.001*	0.402
Interaction	4.57	.01*	0.172	0.42	.66	0.076
PP
Trial	3.47	.04*	0.136	9.26	<.001*	0.296
Electrode	1.76	.19	0.625	0.85	.36	0.453
Interaction	1.45	.24	0.015	1.47	.24	0.018

Statistically significant.

Figure 6.

Grand average ERPs for Ac-PRT (left panel) and Oc-PRT (right panel).

Figure 7.

Group average scalp plots for trials (PM Retrieval, Ongoing, Cue) for ERPs (P2, N2, N4, and PP).

P2 Findings

P2 Peak Latency

A significant main effect of trial for P2 peak latency was observed in both tasks: Ac-PRT (F = 3.160, P = .04) and Oc-PRT (F = 7.55, P < .001). Post hoc tests showed that P2 peak latency was significantly longer for Cue trials compared to Ongoing trials in Oc-PRT (P < .001). No other results were statistically significant.

P2 Mean Amplitude

A significant main effect of trial for P2 mean amplitude was observed in both tasks [Ac-PRT (F = 34.36, P < .001) and Oc-PRT (F = 33.94, P < .001)], with larger P2 mean amplitude for PM Retrieval trials compared to Ongoing trials in Ac-PRT (P < .001) and Oc-PRT (P = .01). Larger P2 mean amplitude was also observed for Cue trials compared to Ongoing trials in both tasks [Ac-PRT (P < .001) and Oc-PRT (P < .001)]. Additionally, larger P2 mean amplitude was observed for Cue trials compared to PM Retrieval trials in Oc-PRT only (P < .001). A significant main effect of electrode cluster was observed in Oc-PRT (F = 4.63, P = .03), but the post hoc comparisons were not significant. No other results were statistically significant.

N2 Findings

N2 Peak Latency

A significant main effect of trial for N2 peak latency was observed in both tasks: Ac-PRT (F = 12.47, P < .001) and Oc-PRT (F = 14.19, P < .001). Post hoc tests for both tasks showed longer N2 peak latency for PM Retrieval trials compared to Ongoing trials in Ac-PRT (P = .01) and Oc-PRT (P = .02), and for Cue trials compared to Ongoing trials in Ac-PRT (P < .001) and Oc-PRT (P < .001). Longer N2 peak latency for Cue trials compared to PM Retrieval trials was observed for both tasks, but was significant only in Oc-PRT (P = .02). No other results were statistically significant.

N2 Mean Amplitude

A significant main effect of trial for N2 mean amplitude was observed in both tasks: Ac-PRT (F = 22.93, P < .001) and Oc-PRT (F = 3.96, P = .02), with larger N2 mean amplitude for Cue trials compared to Ongoing trials in both tasks: Ac-PRT (P < .001) and Oc-PRT (P = .02). Larger N2 mean amplitude was also observed for Cue trials compared to PM Retrieval trials in Ac-PRT only (P = .001). A significant main effect of electrode cluster was observed in Ac-PRT (F = 32.69, P < .001), with larger N2 mean amplitudes at anterofrontal electrodes compared to frontal electrodes (P < .001). The main effects were qualified by a significant trial by electrode cluster interaction in Ac-PRT (F = 4.28, P = .02). Post hoc tests revealed larger N2 mean amplitude for Cue trials compared to Ongoing trials (P < .001) and PM Retrieval trials (P < .001) at anterofrontal electrodes, and larger N2 mean amplitude for Cue (P < .001) and PM Retrieval (P < .001) trials compared to Ongoing trials at frontal electrodes. No other results were statistically significant.

N4 Findings

N4 Peak Latency

A significant main effect of trial for N4 peak latency was observed in Oc-PRT (F = 5.39, P = .01), but not in Ac-PRT (F = 2.52, P = .09). Post hoc tests revealed longer N4 peak latency for Cue trials compared to Ongoing trials in Oc-PRT (P = .007). A significant main effect of electrode cluster was observed in both tasks: Ac-PRT (F = 7.32, P = .01) and Oc-PRT (F = 6.46, P = .01), with longer N4 peak latencies for anterofrontal electrodes compared to frontal electrodes in both tasks: Ac-PRT (P = .001) and Oc-PRT (P = .01). No other results were statistically significant.

N4 Mean Amplitude

A significant main effect of trial for N4 mean amplitude was observed in both tasks: Ac-PRT (F = 36.89, P < .001) and Oc-PRT (F = 30.84, P < .001), with larger N4 mean amplitude for Ongoing trials compared to PM Retrieval trials in Ac-PRT (P < .001) and Oc-PRT (P < .001), and for Ongoing trials compared to Cue trials in both tasks: Ac-PRT (P < .001) and Oc-PRT (P < .001). Larger N4 mean amplitude was also observed for Cue trials compared to PM Retrieval trials in Ac-PRT (P < .001); however, a larger N4 mean amplitude for PM Retrieval trials compared to Cue trials was observed in Oc-PRT (P = .02). A significant main effect of electrode cluster was observed in Oc-PRT (F = 14.79, P < .001), with larger N4 mean amplitude at frontal electrodes compared to anterofrontal electrodes (P = .002). The main effects were qualified by a significant trial by electrode cluster interaction in Ac-PRT (F = 4.57, P = .01). Post hoc tests revealed larger N4 mean amplitude for Ongoing trials compared to PM Retrieval trials (P < .001) and Cue trials (P = .03) at anterofrontal electrodes. Similar patterns were also observed at frontal electrodes with larger N4 mean amplitude for Ongoing trials compared to PM Retrieval (P < .001) and Cue trials (P = .001). Larger N4 mean amplitude was also observed for Cue trials compared to PM Retrieval trials at anterofrontal electrodes (P < .0001). No other results were statistically significant.

PP Findings

PP Peak Latency

A significant main effect of electrode cluster was observed in both tasks: Ac-PRT (F = 36.62, P < .001) and Oc-PRT (F = 18.24, P < .001), with longer PP peak latency at central electrodes compared to parietal electrodes in Ac-PRT (P < .001) and Oc-PRT (P < .001). No other results were statistically significant.

PP Mean Amplitude

A significant main effect of trial for PP mean amplitude was observed in both tasks: Ac-PRT (F = 3.47, P = .04) and Oc-PRT (F = 9.26, P < .001), with larger PP mean amplitude for Cue trials compared to PM Retrieval trials in both tasks: Ac-PRT (P = .03) and Oc-PRT (P = .0005). Post hoc tests also revealed larger PP mean amplitude for Cue trials compared to Ongoing trials in Oc-PRT (P = .003). No other results were statistically significant.

ERP Data: Trial Type Effects Between Tasks

Between task differences in ERP data were analyzed for PM Retrieval and Ongoing trial types individually given the behavioral differences reported earlier between Ac-PRT and Oc-PRT.

PM Retrieval Trials

Analyses of ERP peak latencies during PM Retrieval trials, with electrode cluster as covariate, revealed no significant task-related effects across the P2, N2, N4, and PP components (P > .05). Analyses of ERP mean amplitudes during PM Retrieval, with electrode cluster as a covariate, revealed no significant task effects in N2 component. Only significant findings are discussed below.

P2 Mean Amplitude

A significant main effect of Task was observed (F[1, 69] = 16.52, P <.001) indicating that P2 amplitudes differed significantly between tasks. The covariate effect of electrode cluster was not significant F(1, 69) = 3.22, P = .077. Post hoc comparisons revealed significantly greater P2 amplitudes for Ac-PRT compared to Oc-PRT (P = .0002).

N4 Mean Amplitude

For N4 mean amplitude, significant main effects were found for both Task (F[1, 69] = 12.16, P <.001) and electrode cluster, F(1, 69) = 9.04, P = .004. Post hoc comparisons showed higher N4 amplitude for OC-PRT compared to Ac-PRT (P = .0010). The N4 amplitude variability was partially attributable to differences in electrode clusters.

PP Mean Amplitude

There was a significant main effect of Task (F[1, 69] = 4.74, P = .033), but the covariate effect was not significant (F[1, 69] = 0.62, P = .434). Post hoc comparisons showed significantly greater PP amplitude for Ac-PRT compared to Oc-PRT trials (P = .0354).

Ongoing Trials

Analysis of Ongoing trials with electrode cluster as covariate revealed no significant task effects in peak latency (P2, N2, N4, and PP; P > .05) or mean amplitude (N2, P2, N4; P > .05) except for PP mean amplitude difference.

PP Mean Amplitude

A significant main effect of Task was found (F[1, 69] = 11.07, P = .0014), but the effect of covariate was not significant (F[1, 69] = 2.17, P = .146). Post hoc comparisons showed significantly greater PP amplitudes for Ac-PRT compared to Oc-PRT (P = .0016).

Discussion

This study examined whether ERPs could distinguish neural mechanisms of PM intention retrieval and execution (assessed using PM Retrieval trials) from intention formation (assessed using Cue trials) and intention retention (assessed using Ongoing trials), across 2 novel tasks, Ac-PRT and Oc-PRT, in young adults. The results showed that both tasks effectively distinguished between trial types with 4 major findings: (i) reaction time was longest for the Ongoing trials followed by PM Retrieval trials and Cue trials; (ii) accuracy was highest for Cue trials, followed by Ongoing trials and PM Retrieval trials; (iii) P2, N2, and N4 components differentiated Ongoing and PM Retrieval trials; and (iv) P2, N2, N4, and PP components differentiated Cue from Ongoing and PM Retrieval trials. In addition, 3 task-specific differences emerged: (i) response time was shorter and accuracy was higher in Ac-PRT compared to Oc-PRT for Ongoing and PM Retrieval trials; (ii) P2, N4, and PP mean amplitudes differed across the tasks for PM Retrieval trials; and (iii) PP mean amplitude differed across the tasks for Ongoing trials. No behavioral differences were observed for Cue trials across the tasks.

Behavioral Findings Related to PM Retrieval, Ongoing, and Cue Trials

Within Task Differences

Participants took longer to respond to Ongoing trials compared to PM Retrieval trials in both tasks. Although both Ongoing and PM Retrieval trials required the same level of semantic processing, participants in the Ongoing task not only made semantic judgments but also had to simultaneously retain the PM intention and monitor for the Cue that signaled PM Retrieval trials. The need for additional monitoring during the Ongoing trials may have contributed to longer reaction times,^5,75
-80 consistent with the idea of “PM cost.”⁸¹

Participants responded fastest during Cue trials compared to both Ongoing and PM Retrieval trials, suggesting that encoding an intention was easier than retaining it during the Ongoing task or executing it during PM Retrieval. This was expected, as Cue trials involved empty standalone shapes that were distinct, and easily captured attention. Additionally, unlike the other trial types, Cue trials did not require semantic judgments simplifying processing demand.

As hypothesized, accuracy was lower on PM Retrieval trials compared to Ongoing trials, consistent with the existing literature.^38,82
-84 This suggests that recalling and executing a pre-formed intention is more error-prone than retaining the intention while performing an Ongoing task. The faster reaction times observed during PM Retrieval trials, compared to Ongoing trials, may have contributed to the increased errors, consistent with the speed-accuracy tradeoff.^85
-87 Accuracy for Cue trials was higher than that for Ongoing and PM Retrieval trials, which likely reflects more straightforward processing of Cues trials due to simplicity and distinctiveness compared to Ongoing and PM Retrieval trials.^4,6,53,58,59

Between Task Differences

Participants responded more quickly and accurately in Ac-PRT compared to Oc-PRT for both PM Retrieval and Ongoing trials. Animals are typically more semantically salient, and their biological relevance facilitates faster recognition and decision-making.^88,89 Research suggests some distinct cognitive and neural mechanisms tied to animal judgment compared to object judgment.^90
-92 Alternately, these task-specific differences may be related to variability in the stimuli that constitute the object category compared to the animal category. Unlike the animal stimuli, object stimuli came from 5 subcategories, potentially introducing inherent stimulus variability and preventing the perception of a template-based judgment and thereby needing more prolonged processing leading to longer reaction times and poorer accuracy in Oc-PRT compared to Ac-PRT.^93
-95

ERP Findings Related to PM Retrieval, Ongoing, and Cue Trials

Within Task Differences

Across both the Ac-PRT and Oc-PRT tasks, P2 amplitude was significantly larger for PM Retrieval trials compared to Ongoing trials, consistent with the PM literature.^46,55 Increased P2 amplitude has been observed with maintenance of intentions over time^96,97 and is linked to attentional control.^98,99 We also observed longer N2 latencies for PM Retrieval trials compared to Ongoing trials in both tasks, further supporting a more involved neural processing during PM Retrieval trials compared to Ongoing trials.¹⁰⁰ Given the link between N2 and cognitive control, a longer N2 latency might also reflect a more cautious or deliberate approach to processing the PM Retrieval trials compared to the Ongoing trials.^48,101,102

A larger N400 amplitude was observed during Ongoing trials compared to PM Retrieval trials in both tasks. The N400 component is linked to semantic processing, particularly the integration of meaning and context.^50,103,104 The increased N400 amplitude in Ongoing trials likely reflects not only semantic processing for making a category-level judgment, but also a greater degree of contextual processing, given the demands for monitoring the occurrence of Cue trials or PM Retrieval trials.^41,50

Within both tasks, P2 amplitude, N2 latency, and N2 amplitude differentiated Cue trials from Ongoing trials. The larger P2 and N2 amplitudes, along with the longer N2 latency, were observed for Cue trials compared to Ongoing trials. In early stages of processing, Cue trials appear to draw greater attention due to their simplicity and potentially distinctness, as indicated by the larger P2. Additionally, increased N2 amplitude and longer latency suggest enhanced resource allocation, likely due to higher relevance of these trials for intention encoding, consistent with the literature.^105,106

In the later stages of processing, a larger N400 amplitude was observed for Ongoing trials compared to Cue trials in both tasks. Ongoing trials involve more advanced semantic and contextual processing, consistent with larger N400 amplitudes.^2,107,108 When comparing Cue trials with PM Retrieval trials, a larger PP amplitude was observed for Cue trials, likely reflecting the enhanced resource allocation to prioritize and prepare for subsequent PM retrieval. This finding is consistent with previous PM studies showing that Cue trials are given higher relative importance, as individuals must anticipate and actively prepare for future PM retrieval.^105,106 The Cue might serve as a preparatory signal, enhancing the allocation of cognitive resources to facilitate successful retrieval of the intention during the later PM retrieval phase.⁵⁰

Between Task Differences

Greater PP mean amplitude for ongoing trials was observed in Ac-PRT compared to Oc-PRT. These findings might be related to the fact that biologically relevant stimuli, such as animals, engage greater attentional resources for cognitive evaluation through template-based processing compared to objects.^109,110 PM Retrieval trials in Ac-PRT were also associated with significantly greater P2 mean amplitude, reflecting enhanced early attentional processing. In contrast, PM Retrieval trials in Oc-PRT elicited greater N4 mean amplitude, suggesting increased semantic or contextual processing demands. A greater PP amplitude for PM Retrieval trials in Ac-PRT was also observed, consistent with more robust retrieval-related activity. These findings highlight the influence of task type on both early and late-stage neural processes involved in PM.

While this study provides valuable insights into the neural processes underlying PM retrievals, future research should aim to address the limitations in the current work. The absence of a pure baseline condition, such as a block consisting solely of Ongoing trials without PM demands, limits our ability to fully disentangle general task-related activity from intention-specific neural signatures. Including such a condition in future designs would strengthen interpretations of ERP effects related to intention maintenance and retrieval. A more diverse and representative participant sample estimated using power analysis, particularly with respect to sex distribution, would enhance the generalizability of the findings. In terms of experimental design, incorporating an equal number of Lure trials would enhance the complexity of the task. This would allow for a more robust comparison of ERP responses across trials that share partial properties with the PM Retrieval trials, helping to better differentiate brain activity related to true retrieval from activity associated with similar, but irrelevant, stimuli. While the design of the paradigm in the current study aimed to minimize emotional bias, we acknowledge that the inclusion of animal stimuli typically associated with negative emotional valence (eg, snake, bear, spider) in the ongoing task may have subtly influenced attentional dynamics and arousal. Although these stimuli were limited in frequency and excluded from retrieval trials, their presence could still affect participants’ cognitive processing, particularly in populations with heightened emotional sensitivity, such as older adults. Additionally, while tools (eg, axe, saw) were presented in a neutral, decontextualized format, we recognize that individual differences in emotional associations may still exist. Future studies may consider further controlling for emotional valence or systematically assessing its impact on PM performance. Furthermore, adopting data reduction approaches, such as Principal Component Analysis (PCA), could help identify and isolate electrode clusters that contribute most meaningfully to the observed ERP patterns, enhancing the precision of neural interpretations. While the present study focused on ERPs, exploring event-related spectral perturbations could offer additional insights into the neurophysiological mechanisms underlying PM.

Conclusion

In conclusion, this study established the feasibility of 2 novel PM tasks in examining neural mechanisms liked to PM within the context of semantic judgment. The use of a picture-based PM task embedded within a semantic categorization framework in the current study offers several advantages. Unlike traditional paradigms, such as the Lexical Decision Task (LDT), the current task minimizes the confounds related to literacy and can be used with a broader range of participants. Also, it mirrors real-world PM scenarios, where intentions are often triggered by meaningful visual cues rather than abstract letter strings used in many PM EEG studies. Embedding PM cues within a semantic judgment task which involves basic categorizations, adds to the ecological value, given that semantic categorization is a part of day-to-day functioning. Our findings demonstrate that both tasks effectively differentiated between the distinct phases of PM processing as reflected in ERP (P2, N2, N4, and PP) differences across Cue, Ongoing, and PM Retrieval trials. Additionally, task-specific differences in both behavioral and ERP data highlight the interplay between PM and semantic processing. Overall, this study provides promising evidence for the use of cue-based PM tasks in exploring the neurocognitive underpinnings of PM in aging and clinical populations in future studies.

Footnotes

Acknowledgements

The authors thank Elizabeth Lydon, Teresa Warren, Alexa Worthley, Alexandra Gonzalez, Sneha Nippani, Patricia Jurzyk, and Shansa Viswanathan for their invaluable assistance in data collection. The authors thank James Shim for their assistance in data analysis.

ORCID iDs

Eliza Baby

Shraddha A. Shende

Sally Grace Rogers

Natalia Rzepa

Raksha A. Mudar

Ethical considerations

The study was reviewed and approved by the Office for Protection of Research Subjects University of Illinois Urbana-Champaign (Approval no. IRB24-1702).

Consent to participate

Written informed consent was obtained from all the participants in the study before any study procedures were carried out.

Author contributions

EB: Conceptualization, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing. SS: Writing – review & editing. SGR: Investigation, Visualization. NR: Investigation, Visualization. RAM: Conceptualization, Methodology, Resources, Writing – original draft, Writing – review & editing, Supervision.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The data that support the findings of this study are available upon request and completion of a reliance agreement. R code and output and stimuli will be made available upon reasonable request.

References

Einstein

McDaniel

MA.

Normal aging and prospective memory. J Exp Psychol Learn Mem Cogn. 1990;16(4):717-726.

West

Herndon

Covell

Neural correlates of age-related declines in the formation and realization of delayed intentions. Psychol Aging. 2003;18(3):461-473.

Román-Caballero

Mioni

Time-based and event-based prospective memory in mild cognitive impairment and Alzheimer’s disease patients: a systematic review and meta-analysis. Neuropsychol Rev. 2025;35:102-125. Published online November 14, 2023. doi:10.1007/s11065-023-09626-y

McDaniel

Einstein

GO.

Strategic and automatic processes in prospective memory retrieval: a multiprocess framework. Appl Cogn Psychol. 2000;14(7):S127-S144.

Smith

RE.

The cost of remembering to remember in event-based prospective memory: investigating the capacity demands of delayed intention performance. J Exp Psychol Learn Mem Cogn. 2003;29(3):347-361.

Scullin

McDaniel

Shelton

JT.

The dynamic multiprocess framework: evidence from prospective memory with contextual variability. Cogn Psychol. 2013;67(1-2):55-71.

Cona

Scarpazza

Sartori

Moscovitch

Bisiacchi

PS.

Neural bases of prospective memory: a meta-analysis and the “Attention to Delayed Intention” (AtoDI) model. Neurosci Biobehav Rev. 2015;52:21-37.

Smith

Bayen

UJ.

The effects of working memory resource availability on prospective memory: A formal modeling approach. Exp Psychol. 2005;52(4):243-256.

Wesslein

Rummel

Boywitt

CD.

Differential effects of cue specificity and list length on the prospective and retrospective prospective-memory components. J Cogn Psychol. 2014;26(2):135-146.

10.

Ball

Wiemers

Brewer

GA.

Individual differences in memory and attention processes in prospective remembering. Psychon Bull Rev. 2022;29(3):922-933.

11.

Miyake

Friedman

Emerson

Witzki

Howerter

Wager

TD.

The unity and diversity of executive functions and their contributions to complex “Frontal Lobe” tasks: a latent variable analysis. Cogn Psychol. 2000;41(1):49-100.

12.

Kliegel

Ropeter

Mackinlay

Complex prospective memory in children with ADHD. Child Neuropsychol J Norm Abnorm Dev Child Adolesc. 2006;12(6):407-419.

13.

Altgassen

Schmitz-Hübsch

Kliegel

Event-based prospective memory performance in autism spectrum disorder. J Neurodev Disord. 2010;2(1):2-8.

14.

Gallant

Spaniol

Yang

Age differences in cue utilization during prospective and retrospective memory monitoring. Psychol Aging. 2019;34(4):545-557.

15.

Ball

Peper

Alakbarova

Brewer

Gilbert

SJ.

Individual differences in working memory capacity predict benefits to memory from intention offloading. Mem Hove Engl. 2022;30(2):77-91.

16.

Hunter Ball

Pitães

Brewer

. Individual differences in episodic memory abilities predict successful prospective memory output monitoring. Memory. 2018;26(8):1159-1168.

17.

Burgess

Quayle

Frith

CD.

Brain regions involved in prospective memory as determined by positron emission tomography. Neuropsychologia. 2001;39(6):545-555.

18.

Burgess

Scott

Frith

CD.

The role of the rostral frontal cortex (area 10) in prospective memory: a lateral versus medial dissociation. Neuropsychologia. 2003;41(8):906-918.

19.

Burgess

Dumontheil

Gilbert

SJ.

The gateway hypothesis of rostral prefrontal cortex (area 10) function. Trends Cogn Sci. 2007;11(7):290-298.

20.

Lamichhane

McDaniel

Waldum

Braver

TS.

Age-related changes in neural mechanisms of prospective memory. Cogn Affect Behav Neurosci. 2018;18(5):982-999.

21.

Okuda

Fujii

Ohtake

, et al. Differential involvement of regions of rostral prefrontal cortex (Brodmann area 10) in time- and event-based prospective memory. Int J Psychophysiol. 2007;64(3):233-246.

22.

Zhuang

Kuo

Lin

Yang

Hsu

YH.

Prospective memory and regional functional connectivity in subcortical ischemic vascular disease. Front Aging Neurosci. 2021;13:686040.

23.

Menéndez-Granda

Schmidt

Orth

, et al. The effect of loss incentives on prospective memory in healthy older adults: study protocol of a randomized controlled trial using ultra-high field fMRI. BMC Psychiatry. 2023;23(1):722.

24.

Cona

Santacesaria

Scarpazza

Envisioning the future: an ALE meta-analysis on neural correlates of future thinking, prospective memory and delay discounting. Neurosci Biobehav Rev. 2023;153:105355.

25.

den Ouden

Frith

Blakemore

SJ.

Thinking about intentions. Neuroimage. 2005;28(4):787-796.

26.

Gilbert

Gollwitzer

Cohen

Burgess

Oettingen

Separable brain systems supporting cued versus self-initiated realization of delayed intentions. Learn Mem Cogn. 2009;35(4):905-915.

27.

Burgess

Gonen-Yaacovi

Volle

Functional neuroimaging studies of prospective memory: what have we learnt so far?

Neuropsychologia. 2011;49(8):2246-2257.

28.

Momennejad

Haynes

JD.

Human anterior prefrontal cortex encodes the “what” and “when” of future intentions. Neuroimage. 2012;61(1):139-148.

29.

Burgess

Crum

Pinti

, et al. Prefrontal cortical activation associated with prospective memory while walking around a real-world street environment. Neuroimage. 2022;258:119392.

30.

Vicentin

Cona

Marino

Bisiacchi

Mantini

Arcara

Prestimulus functional connectivity reflects attention orientation in a prospective memory task: A magnetoencephalographic (MEG) study. PLoS One. 2025;20(2):e0319213.

31.

West

Covell

Effects of aging on event-related neural activity related to prospective memory. NeuroReport Rapid Commun Neurosci Res. 2001;12(13):2855-2858.

32.

Zöllig

Martin

Kliegel

Forming intentions successfully: differential compensational mechanisms of adolescents and old adults. Cortex. 2010;46(4):575-589.

33.

Hering

Kliegel

Rendell

Craik

FIM

Rose

NS.

Prospective memory is a key predictor of functional independence in older adults. J Int Neuropsychol Soc JINS. 2018;24(6):640-645.

34.

Hering

Wild-Wall

Falkenstein

, et al. Beyond prospective memory retrieval: encoding and remembering of intentions across the lifespan. Int J Psychophysiol. 2020;147:44-59.

35.

Kliegel

Jäger

Phillips

LH.

Adult age differences in event-based prospective memory: a meta-analysis on the role of focal versus nonfocal cues. Psychol Aging. 2008;23(1):203-208.

36.

Cona

Arcara

Tarantino

Bisiacchi

PS.

Electrophysiological correlates of strategic monitoring in event-based and time-based prospective memory. PLoS One. 2012;7(2):e31659.

37.

Einstein

Holland

McDaniel

Guynn

MJ.

Age-related deficits in prospective memory: the influence of task complexity. Psychol Aging. 1992;7(3):471-478.

38.

Park

Hertzog

Kidder

Morrell

Mayhorn

CB.

Effect of age on event-based and time-based prospective memory. Psychol Aging. 1997;12(2):314-327.

39.

West

McNerney

Travers

Gone but not forgotten: the effects of cancelled intentions on the neural correlates of prospective memory. Int J Psychophysiol. 2007;64(3):215-225.

40.

Czernochowski

Horn

Bayen

UJ.

Does frequency matter? ERP and behavioral correlates of monitoring for rare and frequent prospective memory targets. Neuropsychologia. 2012;50(1):67-76.

41.

Cona

Bisiacchi

Moscovitch

The effects of focal and nonfocal cues on the neural correlates of prospective memory: insights from ERPs. Cereb Cortex. 2014;24(10):2630-2646.

42.

Scolaro

West

Cohen

AL.

The ERP correlates of target checking are dependent upon the defining features of the prospective memory cues. Int J Psychophysiol. 2014;93:298-304.

43.

Guynn

McDaniel

Einstein

GO.

Remembering to perform actions: A different type of memory? In: Zimmer

Cohen

Guynn

Engelkamp

Kormi-Nouri

Foley

, eds. Memory for Action: A Distinct Form of Episodic Memory? Oxford University Press; 2001:25.

44.

Guynn

MJ.

A two-process model of strategic monitoring in event-based prospective memory: Activation/retrieval mode and checking. Int J Psychol. 2003;38(4):245-256.

45.

McDaniel

Einstein

GO.

Prospective Memory: An Overview and Synthesis of an Emerging Field. Sage Publications; 2007.

46.

Brewer

Knight

Marsh

Unsworth

Individual differences in event-based prospective memory: evidence for multiple processes supporting cue detection. Mem Cognit. 2010;38(3):304-311.

47.

West

The temporal dynamics of prospective memory: a review of the ERP and prospective memory literature. Neuropsychologia. 2011;49(8):2233-2245.

48.

Wang

Cao

Cui

Shum

Chan

RC.

The relation between prospective memory and working memory: Evidence from event-related potential data. Psych J. 2013;2(2):113-121.

49.

Wilson

Cutmore

TRH

Wang

Chan

RCK

Shum

DHK

. Effects of cue frequency and repetition on prospective memory: An ERP investigation. Int J Psychophysiol. 2013;90(2):250-257.

50.

Cruz

Miyakoshi

Makeig

Kilborn

Evans

ERPs and their brain sources in perceptual and conceptual prospective memory tasks: commonalities and differences between the two tasks. Neuropsychologia. 2016;91:173-185.

51.

Zuber

Kliegel

Ihle

An individual difference perspective on focal versus nonfocal prospective memory. Mem Cognit. 2016;44(8):1192-1203.

52.

West

Herndon

Crewdson

SJ.

Neural activity associated with the realization of a delayed intention. Brain Res Cogn Brain Res. 2001;12(1):1-9.

53.

West

Ross-Munroe

Neural correlates of the formation and realization of delayed intentions. Cogn Affect Behav Neurosci. 2002;2(2):162-173.

54.

West

Krompinger

Neural correlates of prospective and retrospective memory. Neuropsychologia. 2005;43(3):418-433.

55.

Crook-Rumsey

Howard

Hadjiefthyvoulou

Sumich

Neurophysiological markers of prospective memory and working memory in typical ageing and mild cognitive impairment. Clin Neurophysiol. 2022;133:111-125.

56.

Cousens

Cutmore

Wang

Wilson

Chan

RCK

Shum

DHK

. Effects of perceptual and semantic cues on ERP modulations associated with prospective memory. Int J Psychophysiol. 2015;98(1):151-156.

57.

Zöllig

West

Martin

Altgassen

Lemke

Kliegel

Neural correlates of prospective memory across the lifespan. Neuropsychologia. 2007;45(14):3299-3314.

58.

Kliegel

Ballhausen

Hering

Ihle

Schnitzspahn

Zuber

Prospective memory in older adults: where we are now and what is next. Gerontology. 2016;62(4):459-466.

59.

Cejudo

López-Rojas

Gómez-Ariza

Bajo

MT.

ERP correlates of prospective memory and cue focality in children. Brain Sci. 2022;12(5):533.

60.

Bisiacchi

Schiff

Ciccola

Kliegel

The role of dual-task and task-switch in prospective memory: Behavioural data and neural correlates. Neuropsychologia. 2009;47(5):1362-1373.

61.

Moore

Price

CJ.

A functional neuroimaging study of the variables that generate category-specific object processing differences. Brain. 1999;122 ( Pt 5)(5):943-962.

62.

Noppeney

Price

Penny

Friston

KJ.

Two distinct neural mechanisms for category-selective responses. Cereb Cortex. 2006;16(3):437-445.

63.

Proverbio

Del Zotto

Zani

. The emergence of semantic categorization in early visual processing: ERP indices of animal vs. artifact recognition. BMC Neurosci. 2007;8:24.

64.

Rice

Watson

Hartley

Andrews

TJ.

Low-level image properties of visual objects predict patterns of neural response across category-selective regions of the ventral visual pathway. J Neurosci. 2014;34(26):8837-8844.

65.

Nasreddine

Phillips

Bédirian

, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc. 2005;53(4):695-699.

66.

Snodgrass

Vanderwart

A standardized set of 260 pictures: norms for name agreement, image agreement, familiarity, and visual complexity. J Exp Psychol Hum Learn. 1980;6(2):174-215.

67.

Introducing the CRL International Picture-Naming Project (CRL-IPNP). 2000. ResearchGate. Accessed April 21, 2025. https://www.researchgate.net/publication/237433166_Introducing_the_CRL_International_Picture-Naming_Project_CRL-IPNP

68.

Lopez-Calderon

Luck

SJ.

ERPLAB: an open-source toolbox for the analysis of event-related potentials. Front Hum Neurosci. 2014;8:213.

69.

Delorme

Makeig

EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. J Neurosci Methods. 2004;134(1):9-21.

70.

Rugg MD, Coles MGH, eds. Electrophysiology of Mind: Event-Related Brain Potentials and Cognition . Oxford University Press; 1995:xii, 220.

71.

Olofsson

Nordin

Sequeira

Polich

Affective picture processing: an integrative review of ERP findings. Biol Psychol. 2008;77(3):247-265.

72.

Polich

Updating P300: an integrative theory of P3a and P3b. Clin Neurophysiol. 2007;118(10):2128-2148.

73.

R Core Team. R A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, 2023 References - Scientific Research Publishing. Accessed August 14, 2025. https://www.scirp.org/reference/referencespapers?referenceid=3582659

74.

Fitting Linear Mixed-Effects Models Using lme4. ResearchGate. Published 2015. Accessed April 21, 2025. https://www.researchgate.net/publication/263352347_Fitting_Linear_Mixed-Effects_Models_Using_lme4

75.

Brandimonte

Ferrante

Feresin

Delbello

Dissociating prospective memory from vigilance processes?

Psicológica Rev Metodol Psicol Exp. 2001;22(No 1):97-114.

76.

Einstein

McDaniel

Thomas

, et al. Multiple processes in prospective memory retrieval: factors determining monitoring versus spontaneous retrieval. J Exp Psychol Gen. 2005;134(3):327-342.

77.

McDaniel

Einstein

GO.

The neuropsychology of prospective memory in normal aging: A componential approach. Neuropsychologia. 2011;49(8):2147-2155.

78.

Smith

Hunt

McVay

McConnell

MD.

The cost of event-based prospective memory: Salient target events. J Exp Psychol Learn Mem Cogn. 2007;33(4):734-746.

79.

Strickland

Heathcote

Remington

Loft

Accumulating evidence about what prospective memory costs actually reveal. J Exp Psychol Learn Mem Cogn. 2017;43(10):1616-1629.

80.

Cantarella

Mastroberardino

Bisiacchi

Macaluso

Prospective memory: the combined impact of cognitive load and task focality. Brain Struct Funct. 2023;228(6):1425-1441.

81.

Einstein

McDaniel

MA.

Prospective memory: Multiple retrieval processes. Curr Dir Psychol Sci. 2005;14(6):286-290.

82.

Marsh

Hicks

JL.

Event-based prospective memory and executive control of working memory. J Exp Psychol Learn Mem Cogn. 1998;24(2):336-349.

83.

Einstein

Smith

McDaniel

Shaw

Aging and prospective memory: the influence of increased task demands at encoding and retrieval. Psychol Aging. 1997;12(3):479-488.

84.

Harrison

Mullet

Whiffen

Ousterhout

Einstein

GO.

Prospective memory: effects of divided attention on spontaneous retrieval. Mem Cognit. 2014;42(2):212-224.

85.

Fitts

PM.

Cognitive aspects of information processing: III. Set for speed versus accuracy. J Exp Psychol. 1966;71(6):849-857.

86.

Wickelgren

WA.

Speed-accuracy tradeoff and information processing dynamics. Acta Psychol. 1977;41(1):67-85.

87.

Bogacz

Wagenmakers

Forstmann

Nieuwenhuis

The neural basis of the speed-accuracy tradeoff. Trends Neurosci. 2010;33(1):10-16.

88.

Goñi

Arrondo

Sepulcre

, et al. The semantic organization of the animal category: evidence from semantic verbal fluency and network theory. Cogn Process. 2011;12(2):183-196.

89.

Chittka

Skorupski

Raine

NE.

Speed-accuracy tradeoffs in animal decision making. Trends Ecol Evol. 2009;24(7):400-407.

90.

Bozeat

Ralph

Graham

, et al. A duck with four legs: investigating the structure of conceptual knowledge using picture drawing in semantic dementia. Cogn Neuropsychol. 2003;20:27-47.

91.

Caramazza

Mahon

BZ.

The organization of conceptual knowledge: the evidence from category-specific semantic deficits. Trends Cogn Sci. 2003;7(8):354-361.

92.

Collins

Olson

IR.

Knowledge is power: how conceptual knowledge transforms visual cognition. Psychon Bull Rev. 2014;21(4):843-860.

93.

Selfridge

Neisser

Pattern recognition by machine. Sci Am. 1960;203(2):60-68.

94.

Bruce

Green

Georgeson

, eds. Visual Perception: Physiology, Psychology, & Ecology. 4th ed. Psychology Press; 2003:xii, 483.

95.

Bruce

Georgeson

Green

Georgeson

, eds. Visual Perception: Physiology, Psychology and Ecology. 4th ed. Psychology Press; 2014.

96.

Doyle

Rugg

Wells

A comparison of the electrophysiological effects of formal and repetition priming. Psychophysiology. 1996;33(2):132-147.

97.

Rugg

Nieto-Vegas

Modality-specific effects of immediate word repetition: electrophysiological evidence. Neuroreport. 1999;10(12):2661-2664.

98.

Lijffijt

Lane

Meier

, et al. P50, N100, and P200 sensory gating: relationships with behavioral inhibition, attention, and working memory. Psychophysiology. 2009;46(5):1059-1068.

99.

Wongupparaj

Sumich

Wickens

Kumari

Morris

RG.

Individual differences in working memory and general intelligence indexed by P200 and P300: A latent variable model. Biol Psychol. 2018;139:96-105.

100.

Hockey

Cutmore

Inhibitory control in prospective memory: an event related potential comparison of task-switch and dual task processing. Neuropsychologia. 2021;158:107906.

101.

Botvinick

MM.

Conflict monitoring and decision making: reconciling two perspectives on anterior cingulate function. Cogn Affect Behav Neurosci. 2007;7(4):356-366.

102.

Larson

Clayson

Clawson

Making sense of all the conflict: a theoretical review and critique of conflict-related ERPs. Int J Psychophysiol. 2014;93(3):283-297.

103.

Kutas

Federmeier

KD.

Thirty years and counting: Finding meaning in the N400 component of the event-related brain potential (ERP). Annu Rev Psychol. 2011;62:621-647.

104.

Ghosh Hajra

Liu

Song

Fickling

Cheung

TPL

D’Arcy

RCN.

Multimodal characterization of the semantic N400 response within a rapid evaluation brain vital sign framework. J Transl Med. 2018;16(1):151.

105.

Walter

Meier

How important is importance for prospective memory? A review. Front Psychol. 2014;5:657.

106.

Hering

Rendell

Rose

Schnitzspahn

Kliegel

Prospective memory training in older adults and its relevance for successful aging. Psychol Res. 2014;78:892-904. Published online 2014. doi:10.1007/s00426-014-0566-4

107.

Kutas

Hillyard

SA.

Reading senseless sentences: brain potentials reflect semantic incongruity. Science. 1980;207(4427):203-205.

108.

Van Petten

Luka

. Neural localization of semantic context effects in electromagnetic and hemodynamic studies. Brain Lang. 2006;97(3):279-293.

109.

Proverbio

Adorni

D’Aniello

GE.

250ms to code for action affordance during observation of manipulable objects. Neuropsychologia. 2011;49(9):2711-2717.

110.

Freedman

Assad

JA.

A proposed common neural mechanism for categorization and perceptual decisions. Nat Neurosci. 2011;14(2):143-146.

Investigating Prospective Memory Processes: ERP Evidence from Novel Semantic Judgment Tasks

Abstract

Keywords

Introduction

Method

Participants

Experimental Task

EEG Data Acquisition and Preprocessing

ERP Analysis

Statistical Analysis

Results

Behavioral Data: Trial Type Effects Within Tasks

Reaction Time

Accuracy

Behavioral Data: Trial Type Effects Between Tasks

Reaction Time

Accuracy

ERP Data: Trial Type Effects Within Tasks

P2 Findings

P2 Peak Latency

P2 Mean Amplitude

N2 Findings

N2 Peak Latency

N2 Mean Amplitude

N4 Findings

N4 Peak Latency

N4 Mean Amplitude

PP Findings

PP Peak Latency

PP Mean Amplitude

ERP Data: Trial Type Effects Between Tasks

PM Retrieval Trials

P2 Mean Amplitude

N4 Mean Amplitude

PP Mean Amplitude

Ongoing Trials

PP Mean Amplitude

Discussion

Behavioral Findings Related to PM Retrieval, Ongoing, and Cue Trials

Within Task Differences

Between Task Differences

ERP Findings Related to PM Retrieval, Ongoing, and Cue Trials

Within Task Differences

Between Task Differences

Conclusion

Footnotes

Acknowledgements

ORCID iDs

Ethical considerations

Consent to participate

Author contributions

Funding

Declaration of conflicting interests

Data availability statement

References