Sustaining attention is critical for goal-directed behavior, yet individuals frequently experience lapses that compromise performance. Previous studies have suggested that reaction time (RT) slowing in sustained attention tasks may serve a protective role against lapses, while others interpret this same slowing as a marker of attentional disengagement. To address this contradiction, we tested whether slowing down causally reduces attentional errors using real-time behavioral feedback in go/no-go tasks. In Experiment 1 (N = 30), participants performed a Continuous Performance Task (CPT) consisting of three blocks: baseline, control, and feedback. In the feedback block, participants were instructed to slow down their responses whenever the fixation cross changed color, signaling in real time that their previous responses were too fast. Trial-level analyses revealed that slowing down after feedback significantly reduced the likelihood of attentional errors on subsequent trials, although no significant effects were found at the block level. Given substantial inter-individual variability, Experiment 2 (N = 60) added a cognitive battery alongside the CPT to explore potential correlates. Results showed a reduction in lapses in the feedback condition at the block-level, but no reliable associations were found between feedback benefits and individual differences in cognitive control, working memory, or long-term memory. Taken together, these findings demonstrate that RT slowing have a protective effect against sustained attention lapses in go/no-go tasks. Future research is needed to better characterize which individuals benefit most from this strategic slowing.

Classic memory models proposed that the encoding process involved in visual working memory (VWM) controls the bandwidth of encoding in long-term memory (LTM). Behaviorally, VWM and LTM accuracies are reliably correlated at the behavioral level, raising the question of whether LTM encoding uniquely engages processes that are distinct from VWM encoding. To investigate this, we recorded EEG activity as participants completed recognition memory tasks with set sizes of 32 and 128, far beyond typical VWM capacity. Using interelectrode correlation (IC) analysis, we found that IC patterns reliably predicted individual differences in LTM encoding across both set sizes, indicating a robust, domain-general neural signature. Importantly, this predictive power remained even after controlling for VWM and attentional control performance, suggesting that the model captures variance specific to LTM encoding. Temporally, predictive signals emerged only after stimulus onset and persisted for 500-600 ms. Early and late encoding phases involved distinct network structures, reflecting dynamic neural processes underlying individual differences in LTM encoding. Lastly, we showed that alpha band-passed IC, but not theta or beta band-passed IC, selectively predicted individual differences in LTM performance. Together, our findings reveal a unique and temporally dynamic neural signature that supports individual differences in LTM encoding, independent of general cognitive abilities.

Individuals differ in their ability to sustain attention. However, whether differences in sustained attention reflect differences in processes related to attentional control and working memory or long-term memory (LTM) remains underexplored. In Experiment 1, we conducted an online study (n = 136) measuring participants' sustained attention, attention control and working memory, and LTM. We measured sustained attention with an audiovisual continuous performance task in which participants responded to images while inhibiting responses to infrequent targets; attention control and working memory with flanker, change localization, and Simon tasks; and LTM with recognition and source memory tests. Factor analyses revealed that sustained attention formed a distinct factor from attention control and working memory and LTM. Individual differences in the Sustained attention factor robustly predicted individual differences in LTM and, to a lesser extent, attention control and working memory. In Experiment 2, to test how neural signatures of sustained attention related to attention control and working memory and LTM, we analyzed fMRI functional connectivity patterns collected as 20 participants performed the audiovisual continuous performance task. A pretrained connectome-based model of sustained attention predicted participants' performance on out-of-scanner LTM tasks, but not attention control and working memory tasks. Together, these results suggest that individual differences in sustained attention, although correlated with attention control and working memory, are more closely related to LTM.

Humans possess a remarkable ability to recognize visual objects with high fidelity, supported by complex neural mechanisms underlying memory retrieval. ERP studies have identified two key neural signatures of recognition memory: the parietal old/new effect and the frontal old/new effect. Despite extensive research on these ERP components, the extent to which these components reflect distinct memory processes remains debated. In the present study, we investigated how repetitive learning modulates these ERP components. Participants repeatedly studied a fixed list of 32 real-world images across up to five study-test repetitions while EEG was recorded. In addition, a separate set size 1 condition served as a proxy for working memory. Our results showed that with increased repetitions, the parietal old/new effect exhibited enhanced amplitude and earlier peak latency, reflecting more efficient retrieval of well-learned memories. In contrast, the frontal old/new effect remained unchanged in both amplitude and timing. These findings suggest that the parietal old/new effect is a sensitive neural marker of learning-related changes in long-term memory representations, whereas the frontal effect is less influenced by repetition. In addition, despite similarly high accuracy between the well-practiced set size 32 condition and the set size 1 working memory condition, both parietal and frontal old/new effects peaked significantly earlier for set size 1, suggesting that access to working memory is substantially faster than even well-practiced long-term memory. Together, our results highlight the unique role of the parietal old/new effect, but not the frontal old/new effect, in repetitive learning, despite both components being important for successful recognition of learned visual stimuli.

The amount of information in episodic memory (list length) and the time delay between study and retrieval, including serial position at the beginning (primacy) or at the end (recency) of the list, are among the most prominent determinants of subsequent retrieval success. Although both types of factors are central to theories of memory, they are often confounded in experimental designs, making it difficult to determine whether list-length effects reflect genuine interference from storing more items or time-related processes such as decay. Here, we used electroencephalography (EEG) to examine time-resolved neural signatures of visual recognition memory as a function of list length and delay, focusing on event-related potential (ERP) old/new effects and multivariate EEG decoding. Across two experiments, we observed robust list-length effects on the frontal (FN400) and parietal (LPC) ERP old/new components, as well as on decodability of old and new items. Critically, Experiment 2 employed a delay-controlled design and demonstrated that list-length effects persist when study–test delay is controlled ruling out explanations based on recency confounds. However, serial position modulated neural signals independently producing a primacy advantage that was more prominent in frontal activity. This resulted in dissociable neural signatures: a parietal component selectively sensitive to list length and a frontal component selectively sensitive to primacy. These findings show that list length and delay exert separable influences on neural markers of recognition memory and support the view that list-length and temporal factors (delay and serial position) work via different memory mechanisms.

Memory encoding must accommodate increasingly demanding loads, yet it remains unclear whether the underlying brain state is preserved or adaptively reconfigured as load increases. Here, we recorded EEG from 111 participants performing a visual memory task in which they encoded lists of real images ranging from 1 to 128 items, spanning low to high mnemonic demands, followed by a recognition test. We measured the similarity of set-size-specific interelectrode correlation patterns to a low-load reference. Similarity decreased monotonically as list length increased: patterns at set sizes below 4 remained close to the low-load configuration, whereas set sizes of 8 and above progressively shifted toward a distinct high-load configuration. This transition was captured by a single rotating eigenvector, revealing a low-dimensional trajectory from low- to high-load neural geometry. Smaller rotation angles from the low-load template predicted better memory performance at high loads. Together, these findings provide evidence for adaptive reconfiguration and reveal a structured, low-dimensional neural trajectory through which encoding architecture changes as mnemonic demands increase.

Research suggests that items from similar spatiotemporal contexts are more likely to be retrieved together. Here, we sought to test whether preserving temporal order from encoding to test might lead to a retrieval benefit. Then, we tested the participants using test items that were either presented in the same order as during encoding, or the order of items was different due to it being randomized. Across two experiments, we found that maintaining the same temporal order did not improve overall memory performance in the visual old/new recognition task we used. Next, we show that conditional-response probability plots demonstrate temporal grouping from the study order, but the strength of that grouping was unmodulated by whether the same order or a random order was used during recognition testing. Lastly, we used a two-alternative forced-choice task and again found that presenting old items in their original order, without intervening new items, did not enhance recognition memory performance. Thus, we find that for visually presented common objects tested with recognition, we do not see that the same order boosts participants’ ability to retrieve a sequence of visual objects from memory.

Working memory (WM) tasks often require comparing remembered items to test displays, but little is known about how people selectively remove irrelevant information at test. Across three experiments, we used contralateral delay activity (CDA) to track WM load and examine selective removal. In Experiment 1, CDA amplitudes increased with set size even when only one item was probed, suggesting minimal removal based on spatial location. Experiment 2 ruled out spatial grouping by presenting items sequentially in the same location, yet more items were retained for larger set size. In Experiment 3, however, when items belonged to distinct mnemonic categories, CDA amplitudes at test were reduced, consistent with selective removal based on category relevance. Additionally, P3 old-new effects showed that decision speed and strength were influenced by the number of items maintained. Together, these results suggest that people selectively remove WM contents based on categorical relevance, not spatial cues, enabling more efficient memory-based decisions.

When we try to retrieve a representation from visual long-term memory, there is a chance that we will fail to recall seeing it even though the memory is stored in our brain. Here we show that although mechanisms of explicit memory retrieval are sometimes unable to retrieve stored memories, mechanisms of executive control can quickly query memory and determine if a representation is stored therein. Our findings suggest that the representations stored in human memory that cannot be accessed explicitly at that moment are nonetheless directly accessible by the brain's higher level control mechanisms. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

Decades of research have shown working memory (WM) relies on sustained prefrontal cortical activity and visual extrastriate activity, particularly in the alpha (8 to 12 Hz) frequency range. This alpha activity tracks the spatial location of WM items, even when spatial position is task-irrelevant and no stimulus is currently being presented. Traditional analyses of putative oscillations using bandpass filters, however, conflate oscillations with nonoscillatory aperiodic activity. Here, we reanalyzed seven human electroencephalography visual WM datasets to test the hypothesis that aperiodic activity, which is thought to reflect the relative contributions of excitatory and inhibitory drive-plays a distinct role in visual WM from true alpha oscillations. To do this, we developed a time-resolved spectral parameterization approach to disentangle oscillations from aperiodic activity during WM encoding and maintenance. Across all seven tasks, totaling 112 participants, we captured the representation of spatial location from total alpha power using inverted encoding models (IEMs), replicating traditional analyses. We then trained separate IEMs to estimate the strength of spatial location representation from aperiodic-adjusted alpha (reflecting just the oscillatory component) and aperiodic activity and find that IEM performance improves for aperiodic-adjusted alpha compared to total alpha power that blends the two signals. We also identify a distinct role for aperiodic activity, where IEM performance trained on aperiodic activity is highest during stimulus presentation, but not during the WM maintenance period. Our results emphasize the importance of controlling for aperiodic activity when studying neural oscillations while uncovering a functional role for aperiodic activity in encoding visual WM information.

Individual differences in working memory predict a wide range of cognitive abilities. However, little research has been done on whether working memory continues to predict task performance after repetitive learning. Here, we tested whether working memory ability continued to predict long-term memory (LTM) performance for picture sequences even after participants showed massive learning. In Experiments 1-3, subjects performed a source memory task in which they were presented a sequence of 30 objects shown in one of four quadrants and then were tested on each item's position. We repeated this procedure for five times in Experiment 1 and 12 times in Experiments 2 and 3. Interestingly, we discovered that individual differences in working memory continually predicted LTM accuracy across all repetitions. In Experiment 4, we replicated the stable working memory demands with word pairs. In Experiment 5, we generalized the stable working memory demands model to attentional control abilities. Together, these results suggest that people, instead of relying less on working memory, optimized their working memory and attentional control throughout learning. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

Visual imagery refers to the mental generation of visual representations of stimuli, while visual working memory involves retaining visual information for a short period without external input. Due to the conceptual overlap between these two constructs, successful performance on visual working memory tasks may rely on the use of visual imagery to rehearse items during the retention interval. Consequently, individuals with aphantasia, who lack voluntary visual imagery, may experience difficulties with such tasks. However, prior research has suggested that some individuals with aphantasia might employ non-visual strategies to compensate for this deficit. In two experiments, we examined visual working memory performance in aphantasic and control participants across a range of stimulus types. In Experiment 1, participants completed a change localization task using color squares and complex fractals; in Experiment 2, stimuli included real words, phonologically valid pseudowords, and phonologically invalid pseudowords. Across both experiments, aphantasic participants demonstrated significantly impaired visual working memory compared to controls. Notably, their performance was equally impaired for stimuli that were easily verbalizable (i.e., colors and words) and those that were not (i.e., fractals and pseudowords). Furthermore, individual differences in visual imagery ability, as measured by the Vividness of Visual Imagery Questionnaire (VVIQ), significantly predicted working memory performance across all stimulus types. These findings provide direct evidence for the critical role of visual imagery in supporting visual working memory.

Recording the spiking activity from subcellular compartments of neurons such as axons and dendrites during mouse behavior with 2-photon calcium imaging is increasingly common yet remains challenging due to low signal-to-noise, inaccurate region-of-interest (ROI) identification, movement artifacts, and difficulty in grouping ROIs from the same neuron. To address these issues, we present a computationally efficient preprocessing pipeline for subcellular signal detection, movement artifact identification, and ROI grouping. For subcellular signal detection, we capture the frequency profile of calcium transient dynamics by applying fast Fourier transform (FFT) on smoothed time-series calcium traces collected from axon ROIs. We then apply bandpass filtering methods (e.g., 0.05-0.12 Hz) to select ROIs that contain frequencies that match the power band of transients. To remove motion artifacts from z-plane movement, we apply principal component analysis on all calcium traces and use a bottom-up segmentation change-point detection model on the first principal component. After removing movement artifacts, we further identify calcium transients from noise by analyzing their prominence and duration. Finally, ROIs with high activity correlation are grouped using hierarchical or k-means clustering. Using axon ROIs in the CA1 region, we confirm that both clustering methods effectively determine the optimal number of clusters in pairwise correlation matrices, yielding similar groupings to "ground truth" data. Our approach provides a guideline for standardizing the extraction of physiological signals from subcellular compartments during rodent behavior with 2-photon calcium imaging.

Working memory abilities predict various cognitive abilities, such as gF and source memory, suggesting that these cognitive processes relied on working memory and attentional control resources. When attentional resources were occupied by a secondary task, previous research showed that source memory performance was less impaired than recognition memory, implying that working memory abilities would not influence, or at least exerted much less influence on recognition memory performance than source memory performance. Here, we directly tested if working memory and attentional control differences predict visual recognition memory performance across four experiments (n = 841 in total). Surprisingly, we found that working memory and attentional control predicted source nearly always predicted recognition memory performance as robust as source memory (Exp 1, 3 and 4), except when rapid presentation rate overflowed working memory capacities (Exp 2), refuting our earlier hypothesis. Additionally, source memory and recognition memory, despite of different presentation rates across experiments, remained highly correlated across individuals. Our findings suggested that visual long-term memory generally required working memory and attentional control resources.

Working memory tasks often require comparing remembered visual arrays to test displays, yet little is known about how people edit the contents of working memory at test. Across three experiments, we used contralateral delay activity (CDA) as a neural index of working memory load to examine how memory representations are selectively accessed at test. In Experiment 1, when a single test item was probed, CDA amplitudes increased with larger set sizes, indicating that untested items were still actively maintained, suggesting minimal editing based on spatial location. To test whether this was due to spatial grouping, Experiment 2 presented memory items sequentially in different temporal frames but identical spatial locations. The continued maintenance of all items at test suggested that simple spatial grouping could not explain the lack of editing effect seen in Experiment 1. In Experiment 3, however, when items belonged to distinct mnemonic categories, CDA amplitudes at test were reduced, consistent with selective editing based on category relevance. These findings suggest that working memory editing during retrieval is guided by categorical structure rather than spatial position. Supporting this, analysis of the P3 old-new effect revealed that decision speed and strength were influenced by the number of items maintained at test. Together, our results show that while people do not edit their working memory load based on spatial cues, they edit their working memory based on categorical relevance, allowing for more efficient retrieval of task-relevant information. ### Competing Interest Statement The authors have declared no competing interest.

Vigilance naturally drifts over time, coinciding with marked changes in brain-wide functional magnetic resonance imaging (fMRI) signals. Though the precise origins of these hemodynamic changes are unclear, largely separate lines of research have linked different vigilance levels not only to changes in fMRI signal fluctuation amplitudes and functional connectivity, but also to significant variations in autonomic physiology. These findings raise the possibility that vigilance-related modulations in fMRI signals may arise in part from changes in autonomic physiology and their effects on cerebral hemodynamics. Here, using simultaneous recordings of fMRI, EEG-indexed vigilance, respiration, and pulse oximetry, we investigate how the relationship between autonomic and fMRI signals varies systematically as vigilance gradually drifts. Regression analyses indicated that the strength and extent of fMRI-autonomic covariation increased as vigilance diminished, during both resting state and an auditory vigilance task. Spatiotemporally, autonomic signals exhibited early positive correlations and delayed negative correlations with fMRI signals throughout much of the grey matter, accompanied by late positive correlations in the ventricles and periventricular white matter. Low-frequency EEG power fluctuations also demonstrated state-dependent associations with both fMRI and autonomic signals, with effects in fMRI that partially overlapped with those of peripheral autonomic variations. Functional connectivity between most brain networks strengthened as vigilance decreased, especially during resting-state scans, and removing autonomic variance from fMRI signals largely attenuated this effect. Together, these results demonstrate interactions between vigilance levels, autonomic physiology, and brain hemodynamics, showing that the physiological constituents of fMRI signals vary markedly over vigilance levels and brain regions. These findings contribute to knowledge of human brain physiology and toward the accurate parsing, analysis, and interpretation of fMRI data.