Memory Under Stress: How Post Traumatic Stress Disorder Affects Working Memory in Adults: A Scoping Review

Abstract

Post-Traumatic Stress Disorder (PTSD) is consistently linked to multidimensional working memory (WM) impairments, encompassing deficits in sustained attention, verbal and visuospatial processing, and executive control, with inhibitory dysfunction emerging as a key feature. This scoping review synthesizes evidence from 39 studies examining neurobiological mechanisms, trauma-related factors, genetic and hormonal influences, gender differences, and task-specific variability. Findings indicated that PTSD is associated with altered activation and connectivity in the prefrontal cortex, hippocampus, and related neural networks, often resulting in compensatory but inefficient recruitment patterns. Emotional distraction and comorbidities such as depression, alcohol use, and traumatic brain injury can exacerbate cognitive deficits. Performance impairments are evident across both emotional and neutral WM tasks, with visuospatial and updating processes being particularly vulnerable. Risk factors include chronic trauma exposure, older age, APOE ε4 allele, and the BDNF Val66Met (rs6265) polymorphism, while modulators such as oxytocin, cortisol, and physical activity show potential cognitive benefits under specific conditions. Methodological heterogeneity and limited longitudinal data restrict generalizability. These findings underscore the importance of early screening, targeted cognitive interventions, and inclusion of underrepresented populations to refine prevention and treatment strategies for PTSD-related WM deficits.

1. Introduction

Working memory (WM), a core component of short-term memory, is essential for higher-order cognitive functions such as comprehension, learning, and reasoning. The model proposed by Baddeley and Hitch (Baddeley & Hitch, 1974) posits that WM is a limited-capacity system responsible for the temporary storage and manipulation of information necessary for complex cognitive activities. It is often distinguished from short-term memory, which is primarily concerned with the passive storage of information. In contrast, WM enables the active manipulation of information during tasks such as mental arithmetic or language comprehension. According to this model, the system comprises three core subsystems: (a) the phonological loop, responsible for processing verbal and auditory input, (b) the visuospatial sketchpad, which manages visual and spatial information, and (c) the central executive, a supervisory system that regulates attention, allocates cognitive resources, and coordinates the other components (Goldstein, 2011).

Memory, more broadly, is defined as the process of retaining, retrieving, and using information after the original stimulus is no longer present (Goldstein, 2011). Memory systems are typically divided into explicit (conscious) and implicit (unconscious) forms and are understood to operate across three sequential stages: sensory memory, short-term memory, and long-term memory. Incoming sensory stimuli are initially processed through sensory memory; with sufficient attentional engagement, selected inputs are transferred to short-term memory. Information that undergoes further processing is encoded into long-term memory for sustained storage and future retrieval. These hierarchically organized stages function as an efficient filtering mechanism, preventing cognitive overload and ensuring adaptive functioning (Stangor & Walinga, 2014).

Memory disturbances are common in Post-Traumatic Stress Disorder (PTSD) and are part of its diagnostic criteria, as defined in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) (American Psychiatric Association, 2013) and ICD-11 (International Classification of Diseases [ICD], n.d.). PTSD is a psychiatric and behavioral disorder that can develop following exposure to extremely traumatic or life-threatening events. Such events may include war, natural disasters, sexual assault, intimate partner violence, childhood abuse, or severe accidents. Trauma may be experienced directly or indirectly, and its impact can be either individual or collective. In DSM-5, PTSD belongs to the broader category of Trauma- and Stressor-Related Disorders, which also includes Acute Stress Disorder and Adjustment Disorders (Center for Substance Abuse Treatment (US), 2014). PTSD is defined by four symptom clusters:

(a)

Intrusive re-experiencing (e.g., flashbacks or nightmares);

(b)

Avoidance of trauma-related stimuli;

(c)

Negative alterations in cognition and mood (e.g., memory impairments, emotional numbing);

(d)

Hyperarousal and heightened reactivity (e.g., irritability, sleep disturbances).

Symptoms must persist for more than one month and cause clinically significant distress or impairment. The ICD-11 provides a more concise definition of PTSD, focusing on (a) exposure to a threatening event or series of events, (b) re-experiencing through vivid intrusive memories or dreams, (c) avoidance behaviors, and (d) ongoing hypervigilance. PTSD prevalence is estimated at approximately 8% in women and 4% in men, and the condition can severely impair personal, occupational, and social functioning. Risk factors include prior trauma exposure, pre-existing psychiatric conditions, obsessive personality traits, and biological vulnerability.

The interaction between emotional dysregulation and executive control provides a crucial conceptual framework for understanding WM impairments in PTSD. Influential cognitive models highlight that intrusive trauma memories, attentional biases toward threat cues, and heightened emotional responses can overload cognitive control systems, thereby disrupting WM updating, inhibitory control, and attentional shifting (Ehlers & Clark, 2000). Complementary layered models of emotional–cognitive functioning further propose that attention, regulation, monitoring, and WM interact dynamically within emotionally charged contexts, making individuals with PTSD particularly vulnerable to emotional interference and executive dysfunction during cognitive tasks (Drigas & Papoutsi, 2018).

Neurocognitive research has increasingly focused on the effects of PTSD on memory systems, particularly WM, due to its critical role in regulating behavior, attention, and decision-making (Katsarava et al., 2025). PTSD has been consistently associated with functional and structural alterations in brain regions essential to memory, including the hippocampus and prefrontal cortex (Bremner, 2006; Hinojosa et al., 2024; Pankey et al., 2022). These changes may result in impairments in information encoding, retention, and executive functioning. However, findings across studies remain inconsistent, potentially due to differences in trauma type, symptom severity, gender, comorbid psychiatric conditions, and variability in the WM tasks employed (Ben-Zion et al., 2023; Swain et al., 2023). Despite a growing body of literature examining cognitive dysfunction in PTSD, the specific domain of WM remains underexplored in a systematic and integrative manner. Most prior reviews focus broadly on memory or executive functions without isolating WM as a distinct construct. This review addresses a clear gap by synthesizing evidence specifically on WM impairments in PTSD, highlighting task-specific, neurobiological, and demographic factors that have not been comprehensively evaluated in previous work.

The objective of the present scoping review is to examine how PTSD affects WM in adults, focusing on neuropsychological testing, neurobiological mechanisms, trauma-related factors, biological and genetic predictors, gender differences, and task-related variability, within the context of empirical studies conducted in clinical and research settings. The review aims to synthesize existing findings, identify methodological gaps, and inform future research on the cognitive impact of PTSD.

2. Materials and Methods

2.1. Literature Searches

This scoping review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) 22-item checklist (Tricco et al., 2018). The methodological protocol was developed a priori. Two independent reviewers (A.T. and O.G.) conducted a comprehensive literature search using two major electronic databases, PubMed and Scopus, to identify relevant peer-reviewed articles published from database inception through 7 January 2025. The search strategy employed compound keyword combinations and Boolean operators tailored to each database’s syntax. To ensure the identification of all eligible studies, the reviewers also performed a manual search of the reference lists of included articles. In cases of disagreement, a third reviewer (F.C.) was consulted, and consensus was reached through discussion. The detailed search strategy applied to both databases is presented in

Table 1. Search Strategy.

Database	Search Terms/Query	Filters Applied
PubMed	(post traumatic stress disorder OR PTSD) AND (working memory) AND (adults)	- English language - Humans
Scopus	TITLE-ABS-KEY (((post AND traumatic AND stress AND disorder) OR PTSD) AND working AND memory AND adults) AND (LIMIT-TO (PUBSTAGE, “final”)) AND (LIMIT-TO (DOCTYPE, “ar”)) AND (LIMIT-TO (LANGUAGE, “English”)) AND (LIMIT-TO (EXACTKEYWORD, “Adult”) OR LIMIT-TO (EXACTKEYWORD, “Human”))	- Article type: “ar” (article) - Language: English - Keywords: “Adult” or “Human” - Publication stage: Final

.

2.2. Selection Criteria

The scope of this scoping review was intentionally defined to synthesize evidence on WM in adults with PTSD while avoiding unnecessary heterogeneity introduced by mixed-age samples, pediatric populations, or studies centered on post-intervention effects. Adult samples were selected to ensure comparability across neuropsychological assessments, as developmental and aging trajectories substantially alter WM performance. Similarly, intervention studies were excluded unless they provided baseline cognitive data prior to any manipulation, allowing us to examine naturalistic WM associations without treatment-related confounding. WM was treated as a distinct executive function domain to maintain conceptual clarity and avoid conflating WM-specific mechanisms with broader cognitive or affective constructs. These scope decisions support methodological consistency and reflect the objectives of the scoping review design.

Only original, full-text articles published in English and involving adult participants (≥18 years) were eligible for inclusion. Studies were excluded if they were reviews, meta-analyses, guidelines, commentaries, conference abstracts, unpublished manuscripts, or animal studies. Regarding clinical trials or intervention studies, we applied a targeted exclusion criterion: studies were excluded when their primary aim was to evaluate the effects of an intervention on cognitive or neural outcomes. However, intervention studies were included when they provided eligible baseline WM data prior to any experimental manipulation, ensuring that no post-intervention effects biased the synthesis.

2.3. Data Extraction

Data extraction was performed using a predefined data collection form developed in Microsoft Excel. The following information was recorded: authors, year of publication, study design (cross-sectional/longitudinal), country of study, number of patients diagnosed with PTSD (including gender distribution), age range of patients, presence or absence of a control/comparison group, size and gender distribution of the control/comparison group, age range of the control/comparison group, type of trauma (acute/chronic), trauma/stress assessment instruments, WM assessment tools, main findings related to WM, and the principal conclusion of each study.

2.4. Data Analysis

During the review and processing of the articles, no statistical analysis or meta-analysis was performed due to the high heterogeneity observed among the studies. A quantitative synthesis was not attempted due to the high heterogeneity of the included studies, which involved diverse working memory tasks, scoring procedures, and outcome metrics, preventing meaningful aggregation of results. Consequently, the data were analyzed descriptively.

3. Results

3.1. Database Searches

A total of 424 records were initially retrieved from the two databases (Scopus: 269; PubMed: 155). After removing duplicates, 330 unique studies remained. Following the application of the predefined inclusion and exclusion criteria, all irrelevant studies were excluded. As a result, 205 articles were selected for full-text assessment. Of these, 39 studies met the final inclusion criteria and were included in the review (Figure 1).

3.2. Study Characteristics

The 39 studies included were conducted across different regions and countries. As shown in Figure 2, the majority originated from the United States (n = 20, 51.28%), followed by Australia (n = 5, 12.82%), and the United Kingdom (n = 4, 10.26%). Additional contributions came from Germany (n = 2, 5.13%), and from Bosnia, Brazil, Canada, China, Croatia, Iran, Japan, and Korea, with one study each (n = 1, 2.56%). This geographical distribution indicates that research on the effects of PTSD on WM in adults has been heavily concentrated in high-income, English-speaking countries, particularly the United States. In contrast, representation from non-Western and lower-to-middle-income regions remains limited, underscoring the need for broader international engagement in future research. Of the 39 studies, 33 employed a cross-sectional design, providing insights limited to a single point in time. The remaining six studies used a longitudinal design, allowing for the observation of meaningful changes in WM performance over multiple time points.

3.3. Participants’ Profile

Across the 39 studies included, the number of participants diagnosed with PTSD varied substantially (

Table 2. Summary of Participant Characteristics Across Included Studies.

Category	Subcategory	Number of Studies (n)	Percentage (%)
PTSD Sample Size	1–100 participants	32	82.1%
	101–250 participants	1	2.6%
	251–500 participants	2	5.1%
	>500 participants	4	10.3%
Age Reporting	Age range reported	26	66.7%
	Age range not reported	13	33.3%
Control Group	Matched healthy controls	31	79.5%
	No control group	8	20.5%
Control Group Type	Exposed to trauma (no PTSD diagnosis)	3	7.7% of total
	Non-PTSD clinical group (e.g., orthopedic injury)	1	2.6% of total

). Most studies (n = 32) included PTSD sample sizes between 1 and 100 individuals. One study reported a sample size between 101 and 250 participants, two studies included samples ranging from 251 to 500, and four studies investigated PTSD in populations exceeding 500 participants. The age range of PTSD participants spanned from 28 to 81 years, with most studies focusing on middle-aged adults (40–55 years). However, 13 of the 39 studies did not report participant age ranges, representing a limitation in demographic transparency across the literature.

Regarding the use of reference or control groups, 31 studies compared individuals with PTSD to demographically matched healthy controls, while the remaining 8 studies did not include a control group. Among the included studies, three employed comparison groups exposed to traumatic events who did not develop PTSD, and one study used a control group of individuals with orthopedic injuries. These variations highlight the heterogeneity in study design and undescore the importance of interpreting outcomes within the context of the chosen comparison group.

3.4. Trauma Characteristics and Assessment Tools

Participants in the included studies were exposed to two distinct types of trauma, as defined by the respective methodologies: acute trauma, characterized by a sudden onset and intense symptomatology over a short duration, and chronic trauma, marked by prolonged effects that persist for several months or even years following the initial traumatic event. Most studies (n = 36) focused on chronic trauma, whereas only three included both acute and chronic exposure.

A key feature of this review is the diversity of tools used to assess trauma- and stress-related symptoms. Figure A1 (Appendix A) provides an overview of the clinical instruments employed to evaluate the impact of traumatic experiences.

Across the included studies, trauma and stress were assessed using structured clinical interviews (e.g., CAPS, SCID), standardized self-report instruments (e.g., PCL, BDI), and exposure-specific tools (e.g., Life Events Checklist). Neuropsychological assessment of WM consistently relied on two major categories of instruments: (1) verbal and visuospatial span tasks (e.g., Digit Span, Corsi Block/Spatial Span), and (2) executive WM tasks such as the n-back and complex span paradigms. These tools were the most prevalent across studies and formed the foundation for evaluating immediate memory, updating efficiency, and attentional control in PTSD samples.

3.5. Neuropsychological Assessment Tools

A diverse set of standardized neuropsychological tools was employed across the included studies to evaluate multiple domains of cognitive functioning relevant to WM performance in PTSD populations. Memory-specific assessment instruments (Figure A2, Appendix A) provided quantitative indices of immediate recall, delayed recall, recognition memory, and prospective memory, enabling the detection of both subtle and clinically significant impairments. In addition to memory-focused instruments, several broader neuropsychological measures were used to capture attention, executive function, and processing speed, which are closely interlinked with WM capacity (Figure A3, Appendix A).

Analysis of the data extracted from

Table 3. Review included studies. M: Male; F: Female; N/A: Not Available; C: Cross-sectional; L: Longitudinal; PTSD: post-traumatic stress disorder; WM: working memory.

Author, (Year)	Study Origin	Study Type	Patients N	Patients’ Age	Controls	Controls N	Controls’ Age	Type of Trauma/Stress	Main Findings Related To WM
Toomey et al. (2021)	USA	L	470	Ν/A	yes	524	N/A	chronic	Spouses experiencing mental health conditions such as PTSD and anxiety exhibited notable impairments in attention, WM, and general memory functioning. These cognitive difficulties appeared to be relatively independent of their partners’ deployment status. In contrast, individuals diagnosed with chronic multisymptomatic illness (CMI) demonstrated reduced motor speed, along with a general trend toward diminished verbal memory and attention/WM capacities.
Kim et al. (2024)	Republic of Korea	C	22 (12 M/10 F)	Mean Age: 34.2 ± 13.3 years	yes	22 (12 M/10 F)	Mean Age: 33.6 ± 9.9 years	chronic	Relative to healthy controls (HCs), individuals diagnosed with PTSD exhibited reduced gray matter volume (GMV) in the inferior frontal gyrus (IFG). Moreover, they demonstrated lower accuracy and prolonged reaction times on a face recognition task involving trauma-related distractors. In addition, functional imaging revealed heightened activation in the superior temporal gyrus (STG) during trials with trauma-related distractors. This evidence shows that altered neural activation patterns in PTSD may underlie deficits in cognitive control over ambiguous stimuli and impaired regulation of emotional distraction.
Narita-Ohtaki et al. (2018)	Japan	C	42 F	Age Range: 21–59 years	yes	66 F	Age Range: 20–64 years	chronic	Compared to healthy controls, individuals with PTSD exhibited lower performance across all assessed cognitive domains, including immediate memory, visuospatial construction, language, attention, and delayed memory, as well as the overall score on the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS). Among these domains, memory impairments were the most pronounced. PTSD patients who engaged in regular physical exercise demonstrated better performance in delayed memory tasks compared to their non-exercising counterparts—a difference that remained statistically significant even after adjusting for potential confounding factors in a multiple regression analysis.
Runyan et al. (2022)	USA	C	127 (110 M/17 F)	Age Range: 18–59 years	yes	57 (46 M/11 F)	Age Range: 18–59 years	chronic	The results indicated reduced resting-state functional connectivity (rsFC) between white matter (WM) regions of the ventrolateral prefrontal cortex (vlPFC), lateral premotor cortex, and dorsolateral prefrontal cortex (dlPFC), and areas within the dorsal attention and somatomotor networks in both the mild traumatic brain injury (mTBI) and PTSD groups, relative to healthy controls. Compared to individuals with mTBI, those with PTSD exhibited further reductions in rsFC between the lateral premotor WM seed region and the middle occipital gyrus, as well as between the dlPFC WM seed region and the paracentral lobule. Only connectivity involving the vlPFC was correlated with WM performance across participants.
Biscoe et al. (2024)	UK	C	428	Ν/A	no			chronic	Both PTSD and complex PTSD (CPTSD) were associated with greater impairments in inhibitory control and WM. All PTSD symptom clusters, as well as the disturbances in self-organization (DSO) clusters characteristic of CPTSD, were linked to deficits in both inhibition and WM. Furthermore, consistent evidence indicates that sleep deprivation is also closely associated with WM impairments.
Swick et al. (2017)	USA	C	29 M	no significant differences	yes	29 M	no significant differences	chronic	No significant differences were observed in the degree of WM impairment between verbal and visual stimuli among patients. Combat veterans diagnosed with PTSD demonstrated reduced accuracy on a recent probes WM task in comparison to a mixed control group consisting of both combat-exposed and non-exposed veterans. Importantly, the observed WM deficit in accuracy was consistent across stimulus types, with comparable performance decrements for both verbal and visual conditions.
Flanagan et al. (2018)	USA	C	16 (9 M/7 F)	Ν/A	yes	18 (7 M/11 F)	N/A	chronic	The findings further suggest that administration of a single intranasal dose of oxytocin, compared to placebo, was associated with enhanced performance in the most cognitively demanding WM task condition among participants with PTSD. Oxytocin modulated functional connectivity between the left dorsolateral prefrontal cortex (DLPFC) and the anterior cingulate cortex during the n-back task, resulting in increased connectivity. These results offer novel evidence supporting the potential of oxytocin to ameliorate executive functioning deficits in individuals with PTSD.
Nejati et al. (2018)	Iran	C	30 M	Mean Age: 46.62 years	yes	30 M	Mean Age: 46.62 years	chronic	emotional Crucially, WM performance did not significantly differ between emotional and non-emotional stimuli in terms of either accuracy or reaction time. Individuals with PTSD exhibited impaired WM across both stimulus types. The results imply that WM deficits in PTSD are likely driven by a more general underlying dysfunction, which we propose reflects impaired WM updating mechanisms. Specifically, in PTSD, updating processes may be substantially disrupted, resulting in the accumulation of irrelevant information within WM, ultimately overloading its capacity and hindering the encoding of new material.
Lawrence et al. (2023)	USA	C	1244 (98 F/1146 M)	Average Age: 61.8 years	no			chronic	A significant interaction between APOE ε4 status, PTSD, and age was associated with poorer performance in learning and WM, but not in attention or psychomotor functioning. In contrast, the PTSD-by-age interaction was specifically linked to deficits in attention and psychomotor abilities. The results imply that PTSD may represent a critical target for mitigating age-related cognitive decline, particularly in learning and WM domains, among individuals carrying the APOE ε4 allele.
McDermott et al. (2016)	USA	C	27 M	Ν/A	yes	24 M	N/A	chronic	Our core findings indicate that PTSD is associated with physiological alterations during WM processing, with pronounced effects observed in the right supramarginal gyrus and inferior frontal regions. PTSD-related differences emerged during the early encoding phase, where patients exhibited enhanced alpha oscillatory activity in the right inferior frontal gyrus (IFG) compared to healthy controls. As encoding progressed, these differences extended to the right supramarginal and temporal cortices, and persisted throughout the maintenance phase, with sustained involvement of the right IFG.
Haaland et al. (2016)	UK	L	42 F	Ν/A	No	N/A	N/A	chronic	Stronger pre-treatment learning and memory abilities predicted a more favorable response to PTSD treatment. Our findings further indicate that symptom reduction following a multi-component intervention was associated with improvements in cognitive flexibility, specifically in measures labeled inhibition/switching, even after controlling for changes in depressive symptoms. In contrast, changes in PTSD symptom severity were not significantly related to alterations in learning, memory, or WM performance.
Popescu et al. (2019)	USA	C	35	Ν/A	no	N/A	N/A	chronic	Our principal finding demonstrates that greater severity of PTSD symptoms is negatively correlated with load-related increases in alpha-band power during the retention phase of WM, and is further associated with an elevated rate of false positive responses. These results suggest that enhanced alpha oscillatory amplitude during the retention interval may serve as a neurophysiological mechanism for filtering or “forgetting” weakly activated, irrelevant, or outdated information. Conversely, an impaired ability to appropriately amplify alpha activity in task-relevant cortical regions may reflect deficient pattern separation processes, potentially contributing to fear generalization and the re-experiencing of intrusive traumatic memories.
Sumner et al. (2017)	USA	L	14,029 F	Average Age: 25–42 years	no	N/A	N/A	chronic	Among middle-aged women, elevated PTSD symptom severity was associated with poorer performance on measures of psychomotor speed, attention, learning, and WM. Women with a history of trauma and probable depression, as well as those with probable PTSD in the absence of depression, exhibited lower cognitive performance across composite domains when compared to women without trauma exposure or probable depression. The results imply that PTSD, even in the absence of comorbid depression, may be independently associated with cognitive impairments.
Scott et al. (2016)	USA	C	40	Ν/A	yes	38	Ν/A	chronic/acute	Veterans diagnosed with PTSD demonstrated moderately lower performance on time-based prospective memory (PM) tasks compared to controls, with the majority of errors classified as PM failures. Reduced time-based PM performance was specifically associated with the hyperarousal symptom cluster of PTSD. Within the PTSD group, time-based PM was also related to neuropsychological indices of retrospective memory and executive functioning. However, PTSD remained a significant predictor of impaired PM performance even after accounting for age, retrospective memory, and executive function, indicating an independent contribution of PTSD to PM deficits.
Lawn et al. (2022)	USA	L	10,681 F	Mean age: 64.9 years	no	N/A	N/A	chronic	n this sample of middle-aged civilian women, both PTSD and depression were associated with poorer performance on tasks assessing psychomotor speed, attention, learning, and WM. A history of head injury was also related to PTSD and depression, and when multiple head injuries were reported, this was further associated with greater deficits in learning and WM. This indicates that head injury, PTSD, and depression may jointly contribute to cognitive impairments, particularly in domains involving memory and attentional processing.
Schweizer and Dalgleish (2016)	UK	L	11 M/16 F	Median age: 46 years	yes	12 M/19 F (exp.1) 14 M/45 F (exp.2)	Median age: 23 years (exp.1) Median age: 45 years (exp.2)	chronic	Experiments 1 and 2 confirmed the hypothesis that WM capacity (WMC) is reduced in the presence of emotional, as opposed to neutral, distractors across both student and community samples. Experiment 3 extended these findings to a clinical population: among motor vehicle accident survivors, those with a history of PTSD exhibited selectively diminished WMC when exposed to trauma-related emotional distraction, compared to survivors without PTSD.
Havelka Mestrovic et al. (2020)	Croatia	C	199 M	Median Age: 42 years	yes	116 M	Median Age: 39 years	chronic	Our findings confirmed the presence of WM difficulties and attentional deficits in individuals with PTSD. Among veterans diagnosed with PTSD, carriers of the BDNF Val66Met A allele exhibited lower scores in short-term visual memory tasks, potentially reflecting reduced efficiency in processing visual and spatial information relative to GG homozygotes. Moreover, the impact of the BDNF Val66Met polymorphism on cognitive and executive functioning was moderated by age. Importantly, PTSD diagnosis emerged as a highly significant predictor of both diminished executive performance and poorer visual WM.
Stricker et al. (2015)	USA	C	56 F	Ν/A	yes	53 F	Ν/A	chronic	In our sample of women with PTSD, executive functions—including WM and response inhibition/switching—were poorer compared to healthy controls. Consistent with prior research, greater dorsolateral prefrontal cortex (DLPFC) activation has been linked to more effective suppression of unwanted memories in healthy individuals, potentially serving as a compensatory mechanism in PTSD. As hypothesized, verbal learning and memory were positively associated with executive function performance, independent of processing speed, PTSD symptom severity, and depressive symptoms. Additionally, intelligence quotient (IQ) was related to verbal learning and memory outcomes within the PTSD group.
Bisson Desrochers et al. (2021)	Canada	C	31	Age Range: 18–65 years	yes	30	Age Range: 18–65 years	chronic	Police officers diagnosed with PTSD demonstrated lower cognitive performance compared to healthy controls, particularly in the domains of executive functioning, verbal learning and memory, and lexical access. In individuals with PTSD, memory impairments appear to stem primarily from subtle deficits in the encoding of verbal information, which in turn compromise both learning and retrieval processes. Within the PTSD group, higher levels of intrusion symptoms were associated with reduced executive function, attention, and WM. Furthermore, increased intrusion and avoidance symptom severity was linked to slower information processing speed.
Weber et al. (2005)	Australia	C	10 (7 M/3 F)	Mean Age: 50.8 ± 4.6 years	yes	10 (7 M/3 F)	Mean Age: 47.4 ± 3.3 years	chronic	High-resolution event-related potential (ERP) topography and standardized low-resolution brain electromagnetic tomography (sLORETA) source analysis corroborated previous PET findings of abnormal fronto-parietal activity during WM updating in individuals with PTSD. These abnormalities involved attentional control, WM processes, and interactions with medial temporal regions typically engaged during episodic memory tasks. WM-related activity was found to be negatively associated with intrusion and avoidance symptoms, but positively associated with hyperarousal. These deficits were observed even for trauma-neutral stimuli, suggesting that individuals with PTSD experience pervasive difficulties in attending to and integrating novel information into WM
Moores et al. (2008)	Australia	C	13 (8 M/5 F)	Mean Age: 44.23 ± 9.18 years	yes	12 (7 M/5 F)	Mean Age: 40.41 ± 10.93 years	chronic	The findings underscore that individuals with PTSD exhibit abnormal recruitment of brain regions typically involved in WM updating—such as the dorsolateral prefrontal cortex (DLPFC) and inferior parietal lobule (IPL)—even during WM maintenance operations. In contrast, other regions consistently implicated in the neurobiology of PTSD, including the anterior cingulate cortex (ACC), hippocampus, and brainstem, displayed atypically reduced activation during WM updating. These results suggest that PTSD patients must engage broader cortical areas, including WM updating networks, to perform even basic WM maintenance tasks—possibly reflecting the cognitive symptoms commonly reported in PTSD, such as difficulties in concentration and memory.
Mathew et al. (2022)	USA	C	76	Ν/A	no	N/A	N/A	chronic	The data revealed a significant association between visuospatial WM functioning and the re-experiencing symptom cluster of PTSD. Specifically, the visuospatial WM index accounted for a substantial proportion of the variance in re-experiencing symptoms, suggesting that visuospatial WM capacity serves as a meaningful cognitive correlate of this symptom dimension. Additionally, our findings indicate that repeated trauma exposure was positively associated with poorer visuospatial memory performance, further implicating this domain as sensitive to cumulative traumatic load.
Roberts et al. (2022)	USA	L	12,270 F	Age Range: 50–71 years	no	N/A	N/A	chronic	In this large-scale longitudinal cohort study of trauma-exposed women aged 50 to 71, elevated PTSD symptomatology was associated with greater declines in cognitive functioning over time. Specifically, women with a high number of PTSD symptoms demonstrated greater reductions in learning, WM, psychomotor speed, and attentional capacity compared to those with no symptoms. Even among women with moderate PTSD symptom levels, cognitive decline in learning and WM remained significant. Predictive modeling indicated that women with high PTSD symptom burden experienced approximately twice the rate of decline in learning and WM relative to asymptomatic counterparts.
Flaks et al. (2014)	Brazil	C	81 (PTSD+) 70 (PTSD−)	Age Range: 18–60 years	yes	50	Age Range: 18–60 years	chronic	With regard to visual attentional processing, individuals with PTSD (PTSD+ group) demonstrated poorer performance compared to healthy controls (HC) on the forward spatial span task. Furthermore, the PTSD+ group showed deficits in executive functioning, processing speed, and inhibitory control. These results suggest that individuals with PTSD exhibit marked impairments in sustaining attention over time, particularly when rapid processing and inhibition of concurrent stimuli are required. Such cognitive difficulties may reflect underlying neurobiological alterations associated with PTSD, potentially disrupting concentration, information processing speed, selective attention, resistance to distraction, and inhibitory control.
Schweizer and Dalgleish (2011)	UK	C	39	Age Range: 17–65 years	yes	21	Age Range: 17–65 years	chronic	The results indicated that individuals with a current or past history of PTSD performed worse on the emotional WM capacity (eWMC) task compared to controls when processing valence-neutral content. Group differences were most pronounced in trials involving 5- or 6-sentence sequences, while performance on both shorter (4-sentence) and more complex (7-sentence) trials appeared less sensitive to PTSD-related differences. This indicates that trauma-exposed individuals with any history of PTSD exhibit heightened WM impairments in emotionally salient contexts, particularly under moderate cognitive load, compared to individuals without PTSD history.
Koso and Hansen (2006)	Bosnia, Sweden	C	20 M	mid-40s	yes	20 M	mid-40s	chronic	Our findings provide further evidence that PTSD is associated with impaired sustained attention, as individuals with PTSD demonstrated difficulty maintaining vigilance over time. Even during relatively simple visual scanning tasks, the PTSD group exhibited slower initiation and production of well-defined verbal responses, alongside notable impairments in inhibiting automatic or habitual responses. Additionally, deficits in WM were observed, as reflected by lower performance on the WAIS digit span task.
Shaw et al. (2009)	Australia, Canada	C	13 (8 M/5 F)	Age Range: 30–55 years	yes	12 (7 M/5 F)	Age Range: 28–59 yrs	chronic	In healthy controls, updating and maintenance processes within WM engaged partially overlapping but functionally distinct neural networks. In contrast, individuals with PTSD predominantly recruited a single fronto-parietal network for both processes. This pattern suggests a reduced capacity to flexibly modulate neural activity in response to differing task demands, potentially reflecting impaired cognitive control and neural efficiency in PTSD.
Zhang et al. (2013)	China	C	20	Age Range: 18–40 years	yes	20	Age Range: 20–38 years	chronic	In the presence of negative, as opposed to neutral, distractors, individuals with PTSD exhibited heightened activation in emotion-related brain regions, accompanied by reduced activation in regions associated with WM. Compared to healthy controls, the PTSD group appeared more vulnerable to the interfering effects of negative emotional stimuli, as evidenced by both diminished performance on WM tasks and attenuated engagement of WM-related neural networks.
Sippel et al. (2021)	USA	C	31	Age Range: 18–55 years	yes	21	Age Range: 18–55 years	chronic	PTSD may be associated with impaired social WM (SWM) abilities, particularly under conditions of increased information-processing demands. Hyperactivation of the default mode network (DMN), and specifically the dorsomedial prefrontal cortex (DMPFC), may serve both as a compensatory mechanism and a contributing factor to difficulties in social connectedness. The DMN—especially its dorsomedial subsystem—has been shown to preferentially retrieve socially painful memories. In our findings, individuals with PTSD exhibited hyperactivity in this region during high-load SWM tasks (i.e., four-load trials), highlighting the possible interference of trauma-related social content with goal-directed cognitive processes.
Jelinek et al. (2008)	Germany	C	15	Ν/A	yes	18	Ν/A	chronic/ acute	Results from the ANOVA analysis indicated that performance on the WM task remained stable following the evocation of traumatic memories. No significant changes were observed in either trauma-exposed individuals with PTSD or those without the disorder. These findings offer preliminary evidence that neurocognitive impairment in PTSD may not be a direct consequence of transient psychological activation following trauma recall, suggesting a more stable, underlying cognitive dysfunction.
LaGoy et al. (2021)	USA	C	33	Ν/A	yes	46	Ν/A	chronic	Contrary to initial expectations, increases in slow-wave activity (SWA) were inversely associated with improvements in WM among participants with PTSD. Higher SWA levels were also linked to more severe sleep disturbances in this group. Specifically, greater SWA in frontoparietal regions predicted diminished gains—or even declines—in WM performance. In addition, decreased REM-related theta activity was associated with improved WM performance in the PTSD group, a pattern opposite to that observed in healthy controls. This indicates that REM sleep may serve a different functional role in PTSD, and that non-REM SWA could be related to both daytime cognitive inefficiency and nocturnal behavioral symptoms in affected individuals.
Honzel et al. (2014)	USA	C	18 (17 M/1 F)	Mean Age: 33.5 ± 7.2 years	yes	16 (15 M/1 F)	Mean Age: 36.4 ± 8.6 years	chronic	The results support a dissociation between interference resolution and dual-task performance mechanisms during WM processing. While individuals with PTSD exhibited frontal scalp activation patterns comparable to healthy controls under single-task conditions, they failed to demonstrate the expected shift in scalp topography during dual-task conditions. These findings suggest reduced cortical recruitment—particularly within the frontal and parietal regions—during WM updating when cognitive demands are elevated by a concurrent task, indicating possible impairments in cognitive flexibility and resource allocation in PTSD.
Tian et al. (2014)	USA	C	16 M	Age Range: 21–56 years	yes	16 M	Age Range: 21–56 years	chronic	In the present study, statistical analyses of task-evoked hemodynamic responses revealed significant abnormalities in prefrontal activation among veterans with PTSD compared to healthy controls. Control participants displayed relatively symmetrical activation across both hemispheres during all three phases of the WM task. In contrast, veterans with PTSD exhibited left-lateralized activation in the dorsolateral prefrontal cortex (DLPFC) during the encoding and maintenance phases. More critically, PTSD was associated with marked deactivations during the retrieval phase, particularly in the right prefrontal cortex, suggesting disrupted neural engagement during memory access.
Steudte-Schmiedgen et al. (2014)	USA, Germany	C	24	Ν/A	yes	26 (TC) 30 (HC)	Ν/A	chronic/acute	The central finding of this study was that both individuals with PTSD and trauma-exposed controls (TC) exhibited heightened conflict-driven cognitive control adjustments. This pattern coincided with evidence of altered long-term cortisol secretion in both groups. The increased conflict adaptation observed in these trauma-exposed individuals—who were characterized by lower basal cortisol levels—aligns with previous findings showing less flexible cognitive control under conditions of experimentally elevated salivary cortisol following acute stress. Furthermore, in the TC group, a diminished ability to maintain information in WM was associated with more severe intrusion symptoms, indicating a link between intrusive re-experiencing and WM maintenance deficits.
Veltmeyer et al. (2009)	Australia	C	34 (17 M/17 F)	Average Age: 40.67 ± 11.14 years	yes	136 (68 M/68 F)	Average Age: 40.23 ± 11.33 years	chronic	Patients with PTSD exhibited reduced event-related potential (ERP) amplitudes compared to healthy controls. They also demonstrated longer reaction times, greater reaction time variability, and lower accuracy in target detection. Subgroup analyses revealed that unmedicated PTSD patients exhibited delayed reaction times, while only those receiving selective serotonin reuptake inhibitors (SSRIs) were more likely to miss targets. Importantly, despite exhibiting similar levels of symptom severity, medicated patients displayed impaired WM functioning. Interestingly, reduced ERP amplitudes were associated with fewer clinical symptoms, whereas poorer task performance correlated with greater symptom severity, suggesting a possible dissociation between neural reactivity and behavioral outcomes in PTSD.
Galletly et al. (2001)	Australia	C	18 (10 M/8 F)	Mean Age: 40.6 years	yes	18 (8 M/10 F)	Mean Age: 39.4 years	chronic	The absence of late positive activity over frontal regions during target stimulus processing in individuals with PTSD may reflect persistent deficits in the executive control of attentional and evaluative processes. The presence of such abnormalities during a relatively simple task suggests that individuals with PTSD may be at a significant disadvantage when navigating and responding to more complex sensory environments. Consistent with these neural findings, the PTSD group demonstrated lower accuracy in target detection. Taken together, these results point to impairments in both the short-term storage and executive processing components of WM, aligning with theoretical models of PTSD that emphasize the role of hyperarousal and dysfunction within frontal-subcortical circuits.
DeGutis et al. (2015)	USA	C	18 M	Ν/A	yes	19 M	Ν/A	chronic	Within a cohort of trauma-exposed veterans, PTSD symptoms were found to be specifically associated with deficits in inhibitory control, rather than with generalized executive dysfunction. The overlapping symptomatology between PTSD and depression suggests that these inhibitory control deficits may reflect shared psychological distress common to both disorders, rather than being uniquely tied to core PTSD features such as re-experiencing. These findings further support the growing theoretical perspective that impaired inhibitory control constitutes a fundamental cognitive mechanism underlying post-traumatic symptomatology, possibly mediated by shared neural substrates.
Schindler et al. (2020)	Germany	C	58 (4 M/54 F)	Age Range: 18–65 years	yes	39 TC (7 M/32 F) 45 HC (7 M/38 F)	Age Range: 18–65 years	chronic	Patients with PTSD demonstrated greater conflict adaptation compared to the non-traumatized control (NTC) group, along with reduced autobiographical memory specificity relative to both control groups. The latter finding appears to be a particularly robust cognitive correlate of PTSD. Further analyses examining the predictive value of baseline cognitive functioning for treatment outcomes revealed that performance on the backward digit span task predicted reductions in depressive symptoms following cognitive-behavioral therapy (CBT); however, no other significant predictors were identified.
Yehuda et al. (2007)	USA	C	13 M	Mean Age: 63 years Age Range: 52–81 years	yes	17 M	Mean Age: 63 years Age Range: 52–81 years	chronic	The results indicated that cortisol administration enhanced episodic memory performance across both PTSD and control groups. However, improvements in WM performance were observed only in individuals with PTSD. This preferential enhancement of WM in the PTSD group may reflect an interaction between PTSD symptomatology and age-related factors, as previous findings in younger cohorts showed cortisol-related impairments on similar WM tasks. These results suggest that the cognitive effects of cortisol may be modulated by both psychiatric status and aging, particularly in relation to WM processes.

shows notable convergence in the instruments used to assess WM across the 39 included studies. The most frequently administered WM measure was the Digit Span task (forward and backward), appearing in nine studies, typically as part of broader neuropsychological batteries (e.g., WAIS-based or standalone span measures). Visuospatial span tasks such as the Corsi or Spatial Span were used in seven studies, reflecting the strong relevance of visuospatial WM to PTSD-associated cognitive deficits. N-back paradigms, including both 1-back and 2-back variants, were employed in three to four studies, primarily within neuroimaging or pharmacological designs evaluating WM load–dependent activation patterns. Complex span and updating tasks (e.g., recent-probes paradigms, automated complex span) were used in five studies to capture inhibition and updating efficiency. Additionally, six studies used delayed-response WM tasks, often incorporating emotional distractors to examine interference effects relevant to PTSD. Finally, electrophysiological WM paradigms (EEG/ERP) were used in four studies, providing neurophysiological evidence of altered WM-related activation patterns in PTSD.

Collectively, these instruments offered a multidimensional profile of cognitive functioning, allowing researchers to examine not only core memory processes but also attentional control, cognitive flexibility, and inhibitory regulation. The combined use of memory-targeted and broader cognitive measures reflects a comprehensive approach to neuropsychological assessment, facilitating the identification of PTSD-related deficits across multiple cognitive domains and supporting more nuanced interpretations of WM performance in affected individuals. All study characteristics are presented in Table 3.

4. Discussion

4.1. PTSD and WM Impairments

PTSD is consistently linked to substantial WM impairments across verbal, visuospatial, emotional, and neutral tasks (Narita-Ohtaki et al., 2018; Swick et al., 2017). Neurophysiological evidence indicates reduced cognitive efficiency, reflected in broader and less specialized recruitment of prefrontal and parietal regions (Moores et al., 2008) and diminished ERP amplitudes accompanied by slower and less accurate performance (Veltmeyer et al., 2009). WM deficits worsen with increasing symptom severity, including hypervigilance and re-experiencing (Shaw et al., 2009; Popescu et al., 2019; Stricker et al., 2015). PTSD is also characterized by altered neural processing, such as reduced activation in WM networks under negative distraction (Zhang et al., 2013), atypical fronto-parietal engagement during updating (Weber et al., 2005), and impaired processing under dual-task demands (Honzel et al., 2014). Converging evidence across studies (Galletly et al., 2001; Honzel et al., 2014; Moores et al., 2008; Narita-Ohtaki et al., 2018; Popescu et al., 2019; Scott et al., 2016; Shaw et al., 2009; Stricker et al., 2015; Swick et al., 2017; Veltmeyer et al., 2009; Weber et al., 2005) confirms neurobiologically grounded WM dysfunction that compromises both cognitive control and resistance to emotional distraction.

Τo move beyond descriptive reporting, the present review integrates cognitive, neural, and behavioral findings into a conceptual framework explaining why WM impairments emerge in PTSD. Core deficits—such as inefficient updating (Nejati et al., 2018), inhibitory dysfunction (DeGutis et al., 2015; Biscoe et al., 2024), and reduced cognitive flexibility (Honzel et al., 2014; Shaw et al., 2009)—align with neuroimaging evidence of altered prefrontal activation and connectivity (Moores et al., 2008; Runyan et al., 2022) and compensatory DLPFC–parietal recruitment (McDermott et al., 2016). Heightened limbic reactivity and emotional interference (Kim et al., 2024; Zhang et al., 2013) further modulate these deficits, helping explain why some studies report emotion-specific impairments (Schweizer & Dalgleish, 2011), whereas others find more generalized WM disruptions (Swick et al., 2017). Overall, variability in task design, distractor salience, and cognitive load influences the extent to which WM deficits become evident. This framework highlights disrupted top–down control, impaired filtering during maintenance, and inefficient cortical recruitment as key mechanisms underlying reduced resistance to emotional distraction, visuospatial weaknesses, and diminished prospective memory in PTSD.

A distinction emerges between highly consistent and more heterogeneous WM patterns in PTSD. The most robust findings involve impaired inhibition (DeGutis et al., 2015; Biscoe et al., 2024), increased susceptibility to emotional distraction (Kim et al., 2024; Schweizer & Dalgleish, 2011), and reduced prefrontal efficiency during WM encoding and maintenance (Moores et al., 2008; McDermott et al., 2016). By contrast, visuospatial WM, prospective memory, and compensatory activation profiles show greater variability across studies, with visuospatial deficits pronounced in some samples (Havelka Mestrovic et al., 2020; Mathew et al., 2022) but minimal in others. Such inconsistencies likely reflect differences in task characteristics, trauma history, and sample composition, highlighting the need for further work to delineate domain-specific vulnerabilities.

4.2. Neural Mechanisms and Functional Connectivity Alterations in PTSD

Electrophysiological and neuroimaging modalities, including EEG, MRI, and fMRI, consistently show reduced activity and altered connectivity in key regions such as the prefrontal cortex and hippocampus, with resting-state data showing distinct activation patterns in PTSD compared with controls (Shaw et al., 2009). Patients often rely on a single fronto-parietal network for updating and maintenance (Shaw et al., 2009), display compensatory bilateral frontal engagement during encoding and maintenance (McDermott et al., 2016), and demonstrate weakened connectivity across vlPFC, lateral premotor areas, and DLPFC (Runyan et al., 2022). Additional abnormalities include reduced ERP amplitudes in frontal and parietal regions (Weber et al., 2005), atypical recruitment of DLPFC and IPL with diminished ACC and hippocampal activation (Moores et al., 2008), and increased frontal activation during encoding but marked retrieval-phase deactivation, particularly in the right DLPFC (Tian et al., 2014). Emotional distraction further disrupts these networks, with reduced IFG white matter volume and heightened STG activation (Kim et al., 2024), increased limbic responsivity accompanied by suppression of WM-related regions (Zhang et al., 2013), and altered default-mode connectivity with hyperactivity in the dorsomedial subsystem during social WM tasks (Sippel et al., 2021). Collectively, these findings indicate that PTSD involves widespread neural dysregulation, compensatory but inefficient cortical recruitment, and heightened vulnerability to emotional and social interference, contributing to persistent difficulties in concentration and memory.

Beyond the prefrontal, parietal, and temporal alterations emphasized across the included studies, additional neuroanatomical structures central to PTSD pathophysiology may indirectly contribute to WM impairments through their interactions with executive networks. The amygdala, a core hub for threat detection, exhibits heightened responsivity in PTSD and can disrupt prefrontal regulation during cognitively demanding tasks, thereby exacerbating the emotional interference effects documented in several of the reviewed studies (Kim et al., 2024; Zhang et al., 2013). Brainstem structures, including the locus coeruleus, also play a crucial role through their regulation of arousal and attentional vigilance, processes frequently reported as dysregulated in PTSD populations. Dysfunctions in these regions may further modulate the abnormal functional connectivity patterns observed between dorsolateral prefrontal areas and attentional networks.

Complementary to these circuit-level mechanisms, several neurotransmitter systems implicated in stress and cognitive control—namely serotonergic, noradrenergic, dopaminergic, and cannabinoid signaling—could help explain the variability in WM outcomes across PTSD samples. Altered monoaminergic modulation has been associated with changes in executive function and inhibition, findings that align with the inhibitory control deficits consistently reported in the reviewed studies (e.g., DeGutis et al., 2015; Biscoe et al., 2024). Moreover, the heightened emotional interference observed in PTSD, such as the increased sensitivity to trauma-related distractors described by Kim et al. (2024) and Zhang et al. (2013), aligns with known interactions between limbic structures and neuromodulatory systems regulating arousal and attentional flexibility. Although neurotransmitter activity was not directly assessed in the included studies, these systems provide a broader neurochemical context that complements the neural and behavioral patterns synthesized in this review.

4.3. Impact of Trauma Type and Severity on WM in PTSD

Evidence indicates that trauma type, severity, and exposure context significantly shape WM impairments in PTSD. Childhood abuse, combat-related trauma, and chronic stress are strongly associated with dysfunctional WM patterns, with findings showing poorer n-back performance in individuals with childhood trauma, partially improved through oxytocin-related increases in DLPFC–ACC connectivity (Flanagan et al., 2018) and reduced cortisol levels and adaptability following severe trauma exposure regardless of diagnosis (Steudte-Schmiedgen et al., 2014). Combat-related PTSD demonstrates rsFC patterns resembling mild TBI, with only vlPFC connectivity linked to WM (Runyan et al., 2022), while war veterans also show multidimensional deficits in sustained attention and daily functioning (Koso & Hansen, 2006). Comparisons of Acute Stress Disorder (ASD) and PTSD suggest that WM deficits do not consistently differentiate the two, although trauma-exposed individuals demonstrate altered conflict adaptation and PTSD-specific impairments in time-based prospective memory (Jelinek et al., 2008; Steudte-Schmiedgen et al., 2014; Scott et al., 2016). Complex PTSD (CPTSD) similarly involves widespread WM and inhibitory deficits, with emotional instability emerging as the strongest correlate and additional factors—such as comorbidity, alcohol use, ethnicity, and sleep deprivation—further influencing outcomes (Biscoe et al., 2024). Age also modulates trauma-related cognitive effects, as cortisol improves WM in older but not younger veterans (Yehuda et al., 2007). Overall, trauma characteristics interact with neural systems involved in executive control, with DLPFC functioning and modulators such as oxytocin and cortisol playing key roles under specific conditions.

4.4. Performance Patterns in WM Tasks in PTSD

Research consistently shows that PTSD rarely produces deficits in a single cognitive domain; rather, multiple aspects of cognition are often impaired, as revealed through performance on specialized cognitive tasks. Different task types—verbal, visuospatial, or delayed-response—expose distinct dysfunctions depending on the demands involved, offering both diagnostic and theoretical insights. Swick et al. (2017) reported that verbal and visuospatial WM were equally impaired in PTSD, even without emotional content, with patients making more errors and failing to recall simple recently presented stimuli. In partial contrast, Nejati et al. (2018) found that while accuracy and reaction times did not differ significantly, WM impairments emerged for both emotional and non-emotional stimuli, attributed to deficits in updating processes that overload capacity with irrelevant content. Flaks et al. (2014) observed that PTSD patients performed worse than trauma-exposed and healthy controls on simpler attention and executive tasks, especially visuospatial forward span, processing speed, and inhibitory control, with lower Stroop scores indicating selective attention and distraction resistance difficulties. Mathew et al. (2022) linked lower visuospatial WM scores—measured with Automated Complex Span Tasks—to increased re-experiencing symptoms, suggesting a specific role of visuospatial deficits in PTSD’s intrusive phenomena.

Emotional WM deficits have also been highlighted. Schweizer and Dalgleish (2011) found that PTSD participants performed worse on emotional WM capacity tasks only when trauma-related sentences were presented, particularly at moderate difficulty levels. Kim et al. (2024) confirmed slower reaction times and lower accuracy during delayed-response tasks when emotional content was introduced. All these findings underscore targeted deficits in cognitive subdomains, with heightened vulnerability to emotionally negative stimuli and dysfunctional updating processes undermining recall and encoding. Such patterns point to the need for domain-specific assessments and interventions that address the interplay between emotional interference and WM processes in PTSD, thereby refining both diagnostic accuracy and therapeutic strategies.

4.5. Recovery Potential and Prognostic Indicators for WM in PTSD

Evidence suggests that baseline cognitive functioning shapes recovery potential in PTSD, with early diagnosis and intervention linked to better long-term outcomes. Genetic and biological factors also modulate WM performance: the combination of PTSD, APOE ε4, and older age predicts greater WM and learning decline (Lawrence et al., 2023), whereas the BDNF Val66Met GG genotype appears protective for executive and visuospatial memory, with the A allele associated with poorer performance (Havelka Mestrovic et al., 2020). Therapeutic studies further indicate selective cognitive benefits, as intranasal oxytocin enhances WM and executive function in individuals with childhood trauma (Flanagan et al., 2018) and cortisol improves WM only in older veterans, suggesting age-dependent glucocorticoid effects (Yehuda et al., 2007). Pre-treatment cognitive capacity also influences intervention outcomes: higher baseline memory abilities predict better response to psychotherapy despite limited WM gains (Haaland et al., 2016), CBT appears more effective when baseline WM is higher (Schindler et al., 2020), and physical activity may support delayed memory performance (Narita-Ohtaki et al., 2018). Overall, WM appears relatively stable and resistant to direct therapeutic modification, yet higher pre-treatment capacity and early intervention—particularly in individuals with high-risk genotypes such as APOE ε4 or BDNF Val66Met A—are associated with more favorable clinical trajectories.

Longitudinal designs offer a clear advantage for examining the cognitive consequences of PTSD, revealing how impairments develop, persist, or intensify over time. Evidence from large cohorts shows that higher PTSD symptom levels predict accelerated decline in learning, attention, and WM—approximately double that of non-symptomatic individuals—positioning PTSD as a potential driver of age-related cognitive deterioration (Roberts et al., 2022). Cognitive dysfunction appears early and remains evident even after accounting for comorbid depression, with middle-aged women showing broad deficits across psychomotor speed, attention, learning, and WM (Sumner et al., 2017). The combined effects of PTSD, depression, and head injury lead to more severe impairments, underscoring interactions between psychological and neurological trauma (Lawn et al., 2022). Pre-treatment cognitive capacity predicts therapeutic response, as individuals with stronger baseline learning and memory exhibit better clinical improvement despite limited WM recovery (Haaland et al., 2016). Indirectly affected groups, such as spouses of veterans with PTSD or chronic multisymptom illness, also show reduced attention and WM, highlighting stress spillover effects (Toomey et al., 2021). Finally, emotional interference consistently degrades WM performance in trauma-exposed individuals, particularly when trauma-related distractors are present, suggesting a mechanism through which PTSD disrupts executive control (Schweizer & Dalgleish, 2016).

4.6. Gender, Age, and Occupational Variations in PTSD-Related Cognitive Deficits

Gender, age, and occupational context function as important moderators of working-memory outcomes in PTSD, underscoring the need for cognitively informed assessments that account for demographic and exposure-related vulnerabilities.

Women with PTSD frequently present distinct cognitive profiles, with comorbid depression often emerging as a significant factor. Sumner et al. (2017) found that middle-aged women with PTSD exhibited markedly lower psychomotor speed, attention, and learning/WM scores, even after adjusting for depression and other medical conditions. Lawn et al. (2022) reported similarly poor attention and WM performance in women with PTSD and comorbid TBI, noting that the combination of PTSD, depression, and TBI maximally impaired cognitive functioning. Narita-Ohtaki et al. (2018) demonstrated that women with PTSD scored lower across all RBANS domains—especially verbal memory—but that regular exercise was associated with improved delayed recall, although causality remains unconfirmed. Stricker et al. (2015) identified impaired executive functions, including WM and inhibitory/set-shifting abilities, in female veterans, while Haaland et al. (2016) observed that higher pre-treatment learning and memory scores predicted better therapeutic responses, with post-treatment gains concentrated in executive functions.

Occupationally specific findings reinforce the central role of executive dysfunction. Bisson Desrochers et al. (2021) reported that police officers with PTSD had impaired verbal learning and recall, but intact memory retention, with intrusive symptoms linked to underactivation in frontal regions and related networks. Similar patterns of WM, attention, and executive deficits were found in Bosnian war veterans (Koso & Hansen, 2006) and in spouses of veterans with psychiatric diagnoses, including PTSD and anxiety (Toomey et al., 2021). Collectively, these studies indicate that dysfunctional executive control is a key driver of persistent PTSD symptomatology, regardless of gender or occupation. The evidence underscores that PTSD’s cognitive impact extends beyond psychological distress, affecting long-term brain health. Early diagnosis and intervention appear crucial for women and high-risk groups such as veterans and police officers, as they may prevent or mitigate deficits that otherwise emerge and compound over time.

Overall, evidence from he included studies indicates that gender, age, and occupational background moderate the PTSD–WM relationship. Across multiple studies, women with PTSD showed pronounced deficits in verbal memory, psychomotor speed, attention, and WM, often amplified by comorbid depression or traumatic brain injury (Sumner et al., 2017; Lawn et al., 2022). Female-specific findings also revealed reduced RBANS verbal memory scores and widespread executive dysfunction, including inhibitory and set-shifting impairments (Narita-Ohtaki et al., 2018; Stricker et al., 2015). Age similarly emerged as a key moderator: older veterans and older civilian women demonstrated greater declines in attention, learning, and WM, with genetic factors such as the APOE ε4 allele or BDNF Val66Met A allele further exacerbating age-related vulnerability (Lawrence et al., 2023; Havelka Mestrovic et al., 2020). Occupation-specific data show that police officers with PTSD exhibit marked executive deficits, impaired verbal learning, and reduced frontal activation—patterns similarly observed in war veterans and spouses of veterans facing chronic stress exposure (Bisson Desrochers et al., 2021; Koso & Hansen, 2006; Toomey et al., 2021). Collectively, this evidence demonstrates that female sex is associated with a verbal-memory–centered risk profile, aging contributes to compounded declines shaped by genetic susceptibility, and high-risk occupations exhibit pronounced executive dysfunction arising from cumulative trauma exposure. These moderating factors clarify why WM impairments manifest heterogeneously across PTSD populations and underscore the need for subgroup-specific assessment and intervention approaches.

4.7. Strengths and Limitations of the Study

A notable strength of the present review is the comprehensive search conducted across two major databases, PUBMED and SCOPUS, which yielded 39 eligible articles and provided a detailed overview of the topic. Both cross-sectional and longitudinal studies were included, offering insights into cognitive differences at specific time points and across extended periods. The studies spanned a wide geographic range—from Europe to Japan, the United States, and Australia—ensuring a broad demographic and healthcare representation. This approach facilitated an integrated evaluation of cognitive impacts in patients with diverse backgrounds. Furthermore, the review encompassed a wide range of literature, allowing for a comprehensive synthesis of existing scientific findings that can guide future research.

However, several limitations should be acknowledged. Methodological heterogeneity and variability in patient characteristics prevented statistical analysis or meta-analysis, limiting the generalizability of results. The exclusive inclusion of English-language articles introduced potential bias by omitting case studies and clinical trials published in other languages, which may contain valuable data. In approximately 33% of studies, the age range of participants was not reported, hindering meaningful comparisons across comparable populations. Additionally, the scarcity of longitudinal research limits understanding of the progression of cognitive deficits over time or following therapeutic interventions, highlighting an important gap in the current literature. A key methodological limitation of this scoping review is the absence of a formal critical appraisal of study quality. This omission is consistent with the purpose of scoping reviews, which aim to map the extent, nature, and characteristics of the available literature rather than to evaluate the methodological rigor of included studies. Given the heterogeneity of study designs (e.g., neuroimaging, behavioral tasks, longitudinal cohorts, electrophysiology), the application of a single standardized appraisal tool would not have been appropriate.

4.8. Practical Implications and Future Directions

The evidence synthesized in this review underscores the importance of integrating neuropsychological assessment into the routine evaluation of individuals with PTSD, as cognitive impairments—particularly in WM, sustained attention, and executive control—are both prevalent and functionally disabling. Clinicians should consider that these deficits may persist independently of symptom severity and can be exacerbated by comorbidities such as depression, alcohol use, and traumatic brain injury. Tailored rehabilitation strategies, including cognitive training programs, compensatory memory techniques, and interventions targeting inhibitory control, may improve daily functioning and therapeutic engagement. Furthermore, early screening in high-risk groups (e.g., military veterans, police officers, trauma-exposed women) could facilitate preventive measures and timely intervention before cognitive decline becomes entrenched.

Across the included studies, comorbidities operated through different pathways in shaping the PTSD–WM relationship, highlighting the importance of distinguishing confounders, mediators, and effect modifiers. Depression and generalized anxiety emerged as frequent confounders, as both conditions independently impair attention, psychomotor speed, and verbal WM, as demonstrated in large female cohorts and clinical samples (Sumner et al., 2017; Lawn et al., 2022). Sleep disturbances, hyperarousal, and emotional intrusions functioned as potential mediators, partly explaining how PTSD symptom clusters translate into WM inefficiency—for example, intrusive symptoms predicting WM impairments (Bisson Desrochers et al., 2021) or slow-wave sleep abnormalities affecting WM improvement (LaGoy et al., 2021). Other conditions appeared to operate as effect modifiers, amplifying WM deficits rather than explaining them, such as traumatic brain injury and chronic pain in veterans and police officers (Swick et al., 2017; Koso & Hansen, 2006), genetic factors such as APOE ε4 or BDNF Met allele intensifying age-related cognitive decline (Lawrence et al., 2023; Havelka Mestrovic et al., 2020). Together, these patterns indicate that comorbidities do not play a uniform role in PTSD-related WM deficits; instead, they differentially confound, mediate, or modify outcomes, underscoring the need for nuanced interpretation and subgroup-specific assessment strategies.

Based on the patterns identified across the reviewed studies, several cognitive intervention domains appear especially promising. First, inhibitory control training may directly target one of the most consistent deficits observed in PTSD, particularly given the robust evidence for impaired interference suppression and difficulty filtering irrelevant information. Second, visuospatial WM training, including targeted span and updating protocols, may benefit individuals showing visuospatial vulnerabilities and reduced fronto-parietal efficiency. Third, metacognitive strategy training, emphasizing planning, monitoring, and error awareness could support individuals with broader executive dysfunction and diminished cognitive flexibility. Finally, emotion-regulation–based cognitive rehabilitation, integrating regulation strategies within cognitive tasks, may be particularly effective for patients whose WM performance is compromised by emotional distraction or heightened arousal. Collectively, these intervention pathways offer theoretically informed and clinically relevant avenues for future work aimed at mitigating cognitive impairments associated with PTSD.

From a research perspective, future studies should address current methodological limitations by employing longitudinal designs to map the trajectory of cognitive deficits over time and in response to treatment. There is also a pressing need to evaluate the efficacy of emerging therapeutic approaches—such as neuromodulation techniques, targeted pharmacological agents (e.g., oxytocin, glucocorticoids), and structured exercise regimens—in ameliorating PTSD-related cognitive impairments. Multimodal designs combining neuroimaging, genetic profiling, and ecological cognitive assessments will be essential for elucidating causal mechanisms and guiding precision-based, individualized treatment strategies.

5. Conclusions

This scoping review demonstrates that PTSD in adults is consistently associated with multidimensional WM impairments, encompassing deficits in sustained attention, verbal processing, and executive control, with inhibitory dysfunction emerging as a key feature. Neurobiological evidence highlights altered frontal lobe function, atypical neural connectivity, and maladaptive recruitment patterns, often exacerbated by comorbid factors such as depression, alcohol use, and traumatic brain injury. The cognitive impact varies by gender, age, trauma type, and occupational background, with women and high-risk groups such as veterans and police officers showing pronounced vulnerabilities. Although emerging findings suggest potential modulators—including genetic markers, hormonal influences, and lifestyle factors—the heterogeneity of study designs, limited longitudinal data, and language restrictions constrain generalizability. Addressing these gaps through diverse, methodologically robust, and long-term studies will be essential for clarifying causal mechanisms, refining interventions, and ultimately improving cognitive outcomes and quality of life for individuals with PTSD.