Exploring the influence of pregnancy on cognitive function in women: a systematic review

Abstract

Background

Pregnancy has been increasingly recognized for its potential impact on cognitive function influenced significantly by hormonal fluctuations such as estrogen and progesterone. However, the findings from research in this area remain debated, often varying with individual factors and pregnancy trimesters.

Objective

This study aims to systematically review existing literature and empirical research to better understand the phenomenon known as “pregnancy brain” and its association with cognitive change.

Methods

We conducted a systematic review of peer-reviewed English language articles sourced from Medline, Cochrane Library, and PubMed. Using search terms (pregnancy OR pregnant) AND (cognit* OR memory OR cognitive function*), we identified relevant articles published from the inception of these databases to July 2023. Due to the diverse outcomes in identified studies, a meta-analysis was not feasible, and therefore, a narrative synthesis was conducted to summarize findings.

Results

Screening 3892 studies yielded 31 that met inclusion criteria, encompassing 1596 pregnant women and 1450 non-pregnant controls. Notably, seven studies originated from Australia, of which four were rated as low quality. The aggregated findings suggest that pregnancy modestly affects verbal memory and attention compared to non-pregnant controls, with indications that this decline persists postnatally.

Conclusion

Future efforts should prioritize uncovering the underlying mechanisms driving these cognitive changes during pregnancy and consider developing targeted interventions. Such initiatives aim to mitigate potential impacts on maternal well-being, advocating for comprehensive support systems tailored to pregnant individuals navigating cognitive shifts during this transformative period.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-025-07181-3.

Introduction

“Pregnancy brain,” often referred to as “baby brain,” is a common term used to describe potential memory lapses and forgetfulness experienced by pregnant women [1, 2]. This phenomenon is widely acknowledged and discussed in popular pregnancy guides and has even made its way into light-hearted literature [3]. Pregnancy triggers various adaptive changes in maternal physiology and brain function due to hormonal fluctuations, influencing mood and behavior [4, 5]. The impact of pregnancy hormones and neurotransmitters on brain function suggests that higher cognitive functions may also be affected [6].

In the past 25 years, there has been a notable increase in scientific investigation into pregnancy-related cognitive decline [7]. A meta-analysis by Davies et al. demonstrated a significant decline in objective cognitive function in pregnant women compared to non-pregnant women [8]. Neuroimaging studies have also revealed significant structural changes in the maternal brain during and after pregnancy [1]. For instance, Oatridge et al. used MRI to show a significant decrease in brain volume during pregnancy, alongside an increase in lateral ventricle size [9].

Memory is one of the complex brain function that enables the storage and retrieval of information. In humans, the collection of life experiences shapes our identity and sense of self [10]. Long-term memory, including procedural and declarative memory, serves as an extensive repository of knowledge and a record of past experiences. Procedural memory include activities learned through practice and repeated exposure to specific motor tasks. However, declarative memory, which can be consciously recalled, is divided into two categories, semantic memory and episodic memory. Semantic memory involves discrete facts, while episodic memory captures explicit personal experiences, such as a memorable birthday or a wedding day [11]. According to various theoretical perspectives, it is universally acknowledged that every individual possesses a substantial, imperfect and incomplete, collection of long-term memories [11]. Short-term memory is associated with primary memory. Similar to Atkinson and Shiffrin’s view, I believe it represents the mental capabilities that can temporarily retain a limited amount of information in an easily accessible form [11]. Working memory is not entirely separate from short-term memory. It refers to the type of memory utilized for planning and executing actions [11].

Despite these findings, the underlying cause of impaired cognition, especially memory, during pregnancy remains unclear [5]. Research by Hoekzema et al. found that pregnancy is associated with gray matter decline in brain areas involved in social cognition, lasting at least two years postpartum, possibly due to synaptic pruning for specialization in childcare [12]. This social cognitive development may enhance mother-infant relationships and the mother’s ability to identify her children’s needs [5]. Brett and Baxendale reported that 50–80% of pregnant women experience forgetfulness and poor memory [13]. However, these reported impairments may not necessarily reflect objective changes in memory performance [7]. For instance, Cook and Marsiske found that subjective assessments of memory ability were not significantly correlated with actual memory performance [14]. Factors such as personality traits, emotional states, and health conditions like neuroticism, anxiety, and depression may contribute to subjective memory [15, 16].

Until recently, most evidence concerning pregnancy-related memory problems was anecdotal and relied on subjective reports [17, 18]. A significant body of empirical research has now accumulated to address the extent to which self-perceived memory changes reflect actual memory changes [7]. While many studies support the idea that memory is negatively impacted during pregnancy [6, 19–23], others have found no significant effects [24–26]. De Groot summarizes that the evidence for objective cognitive changes during pregnancy is “equivocal.” Given these mixed findings and the growing body of research, it is essential to comprehensively review the available literature to clarify the nature and extent of pregnancy-related cognitive changes [27]. The objective of this study is to evaluate the phenomenon of “pregnancy brain” through a review of existing literature and empirical research, aiming to understand the association between pregnancy and cognitive change.

Methods

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) recommendations were followed when conducting this systematic review [28]. As this is a systematic review, there was no need for Ethics Committee approval.

We sought to summarize the research supporting the link between pregnancy and cognitive change in this review. In an effort to frame the research question, we used the PICO framework (population, intervention (exposure), comparator, and outcome) as follows: Are pregnant women (the population) more likely than non-pregnant controls (the comparator) to experience cognitive change (the outcome) during pregnancy (intervention)?

Search strategy and databases

A comprehensive search for relevant studies was conducted across diverse scholarly sources, encompassing MEDLINE, PubMed, and Cochrane Library from the inception of the respective databases until 31 July 2023. The search strategy was strategically organized using a set of keywords, including ‘memory,’ ‘cognitive function*,’ ‘cognit*,’ ‘pregnant,’ and ‘pregnancy.’ These keywords were combined using Boolean operators to refine the search process and retrieve pertinent articles. To manage the accumulated references, ZOTERO software was employed.

Study selection and eligibility criteria

Our review included all comparative/observational studies (including case-control, cohort, and cross-sectional studies), as well as Randomized Controlled Trials (RCTs) and clinical trials that provided quantitative measures of the relationship between pregnancy and cognitive change. Studies comparing pregnant women to a control group for cognitive changes (enhancement or impairment) were considered eligible. Our search specifically encompassed human studies and was confined to the English language. No constraints were imposed on the year of publication, allowing for the inclusion of studies spanning various time frames. In our search, we excluded study designs such as protocols, conference proceedings, books, and review articles, as these sources do not constitute original empirical research and are not suitable for addressing our specific research question. Furthermore, studies involving non-human subjects, people under the age of 18, and people who already had neurological disorders were not included in the review.

Data extraction

All references were downloaded into Zotero from online databases. The two reviewers (JY and MB) independently identified studies eligible for inclusion after removing duplicates. Data extraction was also carried out independently following full-text screening by the same reviewers. To enable comparison and analysis across studies, the characteristics of each study were summed up and tabulated. Basic study information (last author’s name, publication year), study design, fundamental sample characteristics (total sample size for each studied group separately, maternal ages of both groups), cognitive measures and their domains, and results were the categories into which the information was divided. Only systematic synthesis was done because statistical combination of studies was not possible due to the variation in study components.

Methodological quality assessment

The methodological quality assessment of the included observational studies was performed using the Newcastle-Ottawa scale (NOS), a tool designed to evaluate the quality of nonrandomized studies [29]. Studies are evaluated based on three main criteria: the selection and comparability of the study groups, the ascertainment of the outcome of interest for cohort studies and cross-sectional studies, and the data quality section. Consequently, a cohort study may receive up to three stars for the Selection and Comparability section, five stars for the Outcome section, and five stars for the Data Quality section. A cross sectional study may receive up to three stars for the Selection and Comparability section, five stars for the Outcome section, and only one star for the Data Quality section. Therefore, according to the modified NOS scale, high-quality cohort studies can receive up to thirteen stars, while high-quality cross-sectional studies can only receive up to nine stars. Studies are categorized as “high-quality” if they have seven or more stars for a cohort study or five or more stars for a cross-sectional study. The age and education of the groups under study were the main determinants of comparability. Furthermore, it was thought that for cohort studies, outcomes could be observed after 12 weeks or more of follow-up.

For one study which was a controlled clinical trial we used Methodological Index For Non- Randomized Studies (MINORS) [30]. MINORS is a reliable tool for evaluating the methodological quality of surgical studies that are non-randomized. The scale comprises 11 items, each scored as follows: 0 (not reported), 1 (reported but inadequate), and 2 (reported and adequate). Studies are classified as ‘high-quality’ if they receive 14 or more stars. The total ideal score is 16 for non-comparative studies and 24 for comparative studies. The assessment was performed independently by two reviewers. Any discrepancies in the assessment were resolved through discussion and consensus among the reviewers.

Results

Study selection

The initial electronic searches yielded a total of 3892 studies, supplemented by an addi1tional 18 records sourced from the meta-analysis by Davies et al. [8]. After the exclusion of duplicate and irrelevant articles, 2763 studies remained for further consideration. Among these, 2744 articles were deemed ineligible based on their failure to meet the established inclusion criteria. Subsequently, 19 abstracts were subjected to a more comprehensive evaluation due to their potential alignment with the research objectives. To undertake a thorough examination, the full-text versions of all 19 selected studies were reviewed. Among these, 13 studies satisfied all the predetermined inclusion criteria and were consequently subjected to rigorous analysis. The entire set of analytical procedures was uniformly applied to both the original 13 studies and the supplementary 18 studies. In terms of study design categorization, the analysis unveiled the presence of diverse research methodologies. To visually represent the progression of study selection, the detailed flowchart outlining the selection process is visually depicted in Fig. 1.

Fig. 1.

PRISMA flow chart diagram

Study characteristics

The 31 studies included in the review were classified as 12 cohort studies [2, 20, 22, 27, 31–38], 18 cross-sectional [21, 23–26, 39–51], and one controlled clinical trial study [52]. The majority of studies were conducted in Australia (seven of them), three in the United Kingdom, three in Canada, two in Netherlands, two in USA, two in England, and one study each in New Zealand, Italy, South Africa, and Poland. The settings of the other eight studies weren’t reported. All studies were published in the English language in peer-reviewed journals. Included studies showed a total of 1596 pregnant women and 1450 non pregnant controls. Sample sizes ranged from 20 [22] to 806 [2]. Cognitive functions were assessed by different tasks addressing memory, attention and general cognitive functions. Mostly, WAIS Digit Span (forward and backward) for assessing memory were used [24, 31, 35, 38, 39, 44, 45] as well as Rey AVLT [35, 39, 47, 48]. Two other tests : WAIS III Verbal Paired Associates and WAIS III Logical Memory tests were frequently used [35, 47, 48]. On the other hand, tasks were heterogeneous for attention, with Stroop Color Word task being the most used task [20, 32, 40, 43]. A detailed summary of the study characteristics is available in Supplementary Table 1.

Risk of bias

The quality of cohort and cross sectional studies were assessed by the New Castle Ottawa Scale NOS [29] and the quality of one non randomized clinical trial was assessed using Methodological Index of Non Randomized Studies (MINORS) [30]. Studies with scores ≥ 7 for cohort studies and studies ≥ 5 for cross sectional studies were classified as high-quality studies. For the only clinical trial, a score ≥ 14 was classified as “high quality” study. [2, 20–25, 27, 31–35, 37–40, 43–51] those studies were considered of high quality, while [26, 36, 41, 42] studies were considered of low-quality. Results of the quality assessment of all studies using the NOS and new MINORS scales are summarized in Tables 1 and 2 respectively.

Table 1.

Quality assessment of observational studies

Study	Study design	Selection and comparability	Outcome	Data quality	Score	Quality
De Groot et al. [40]	Cross sectional	0	3	1	4/9	LOW
Anderson et al. [41]	Cross sectional	1	3	0	4/9	LOW
Sharp et al. [23]	Cohort	3	4	0	7/9	HIGH
Bae et al. [29]	Cohort	1	5	3	9/13	HIGH
Casey et al. [24]	Cross sectional	2	5	0	7/9	HIGH
Christensen et al. [2]	Cohort	2	3	3	8/13	HIGH
Glynn et al. [38]	Cross sectional	1	4	0	5/9	HIGH
JT Condon et al. [20]	Cohort	3	5	3	11/13	HIGH
Rendell et al. [37]	Cross sectional	0	5	4	9/13	HIGH
Cuttler et al. [39]	Cross sectional	0	5	0	5/9	HIGH
Casey et al. [31]	Cross sectional	1	5	2	8/13	HIGH
Vanston et al. [36]	Cross sectional	3	5	4	12/13	HIGH
Crawley et al. [32]	Cohort	0	5	4	9/13	HIGH
Farrar et al. [33]	Cohort	2	5	5	12/13	HIGH
Crawley et al. [43]	Cohort	3	5	1	9/9	HIGH
Buckwalter et al. [19]	Cohort	0	3	4	7/13	HIGH
De Groot et al. [21]	Cohort	1	5	2	8/13	HIGH
Christensen et al.[25]	Cross sectional	0	2	1	3/9	LOW
Janes et al. [44]	Cohort	1	5	1	7/9	HIGH
Slim et al. [30]	Cross sectional	0	5	3	8/13	HIGH
Brindle et al. [42]	Cross sectional	2	5	1	8/9	HIGH
Onyper et al. [45]	Cohort	1	5	1	7/9	HIGH
Henry et al. [35]	Clinical Trial	0	5	4	9/13	HIGH
Spataro et al. [50]	Cross sectional	2	4	0	6/9	HIGH
Keenan et al. [22]	Cross sectional	1	5	0	6/9	HIGH
Wołyńczyk-Gmaj et al. [49]	Cross sectional	3	4	0	7/9	HIGH
Harris et al. [34]	Cross sectional	0	3	2	5/13	LOW
Raz et al. [46]	Cohort	0	5	0	5/9	HIGH
Wilson et al. [47]	Cross sectional	0	5	0	5/9	HIGH
Wilson et al. [48]	Cross sectional	1	5	1	7/9	HIGH

Table 2.

Quality assessment of clinical study

Study	A clearly stated aim	Inclusion of consecutive patients	Prospective collection of data	Endpoints appropriate to the aim of the study	Unbiased assessment of the study endpoint	Follow-up period appropriate to the aim of the study	Loss to follow up less than 5%	Prospective calculation of the study size
[53]	2	2	2	2	0	0	2	2
Additional criteria in the case of comparative study
	An adequate control group	Contemporary groups	Baseline equivalence of groups	Adequate statistical analyses	Score	Quality
	2	2	0	2	18/24	HIGH

Findings on cognitive function changes

Assessments of memory impairment during pregnancy

It’s crucial to try and identify any real deficit using rigorous and objective measures of function because there are still frequent subjective reports of memory impairment. Despite 25 years of research, the results remain inconclusive. While some studies such as [22, 23]) report decreased antepartum performance on specific memory tasks, other studies such as [25, 26, 32, 44]) found no evidence of such impairment. The potential domain specificity of memory effects may explain some, but not all, of the variations in results across studies. We will further elucidate these findings by presenting results categorized by the type of memory. The summary of the subtypes of memory are presented in Table 3.

Table 3.

A summary of the results associated with each category of memory

Memory type	Result
Short term Memory	No influence of pregnancy on this type of memory
Working Memory	The backward digit span task did not find any significant declines in working memory, other research using different tasks found working memory performance variations
Long term Memory -Verbal Episodic Memory	Verbal recall and verbal recognition are not affected during pregnancy
Long term memory- Implicit memory	No influence of pregnancy on this type of memory
Visuospatial Memory	No influence of pregnancy on this type of memory
Verbal Auditory Memory	The verbal auditory memory is affected during pregnancy
Prospective Memory	Prospective memory is affected during pregnancy

Short-term memory

Short term memory, is responsible for temporarily holding and processing information needed for cognitive tasks. It’s one of the stages of memory processing and plays a vital role in various cognitive tasks. The following studies [24, 31, 35, 45] made use of the Forward Digit Span Test to gauge this type of memory. None of the four studies mentioned had provided any evidence that there were any differences between the groups in how well they performed on the forward digit span. This finding implies, at least based on the reported findings, that in the context of these studies, that pregnancy did not appear to influence performance on the forward digit span test and thus on short term memory.

Working memory

Working memory and short-term memory are related but distinct components of the memory system, and they differ primarily in terms of their functions, processes, and duration. Working memory is more active, with manipulating and processing information for cognitive tasks, problem-solving, decision-making, and complex thinking. It also allows mental calculations and the following directions. Working memory has a slightly longer duration, typically a few seconds to a couple of minutes, but involves active manipulation and processing of stored information. Its capacity is limited, typically holding around 5–9 items of information, but it can manage complex information by reorganizing or grouping it effectively. In summary, working memory is an active cognitive system that supports higher-level cognitive functions. Recent theories suggest that, at least in a subset of women (who may be depressed), memory functions with a strong executive component, such as working memory, may experience modest declines during pregnancy [53].

Backward digit span was primarily used to gauge working memory; a cognitive system responsible for temporarily holding and manipulating information needed for various mental tasks. This task requires working memory because it requires repeating digit sequences out loud in the opposite order from how they are heard. According to our search, nine studies [2, 24, 31, 35, 36, 38, 39, 44, 45] have made use of it. The backward digit span has typically been used as a control task in studies of memory during pregnancy, administered along with the forward span. All of the studies –with the exception of two [2, 44]-reported no evidence of working memory decline. According to the longitudinal study by Christensen et al., those who were pregnant experienced a greater performance decline than those who remained non-pregnant (p = 0.037); however, after accounting for multiple testing, this difference was no longer significant.

Alternatives of backward digit span have been employed through cognitive neuroscience literature. These alternatives mainly arise because of their higher sensitivity compared to classic backwards digit span. For example, a study by Farrar et al. reported that the pregnant group scored significantly lower than the control group on the Spatial Recognition Memory (SRM) test at the second trimester and postpartum assessments (p = 0.004) [33]. Pregnant women also scored lower overall on the OSPAN test for letters remembered, according to a different study cited by Dorota et al. indicating a decline in working memory performance of pregnant women in their third trimester compared to non-pregnant controls [49]. In these studies, decreased working memory scores were evident in the second or third trimesters of pregnancy. Other studies [24, 31, 35, 44, 45] using other backwards digit span alternatives such as verbal working memory task, modified reading span task, Letter Number Sequence, spatial span backwards, reading span tests didn’t show any significant differences in performance of pregnant women in different trimesters compared to their controls. One study [36] did compare for the effect of pregnancy on working memory but according to fetal sex. The results of the study showed a significant effect of fetal sex on maternal cognition. Women pregnant with boys consistently outperformed women pregnant with girls on difficult tests of working memory and spatial ability.

In aggregate, there are conflicting results from the research on working memory during pregnancy. While some research using the backward digit span task did not find any significant declines in working memory, other research using different tasks found that working memory performance varied throughout the various trimesters of pregnancy.

Long-term memory

Long-Term Memory involves the storage of information for an extended period, ranging from minutes to a lifetime. It is further divided into several subtypes:

Verbal Episodic Memory

The ability to store and recall details of past experiences, events, and knowledge in verbal or linguistic form is known as verbal episodic memory. A key part of long-term memory, verbal episodic memory is essential to our capacity to recall and communicate our life stories and personal histories through language and communication. It has been the subject of the majority of objective studies on cognitive changes during pregnancy. By using word list learning or paragraph recall tasks to measure memory, verbal episodic memory in particular has received considerable attention. In these studies, recall methods—either free recall or cued recall —are commonly used to assess memory. Memory is assessed immediately or after a delay or both. The use of recognition methods is uncommon (for instance [23, 35]), .

Verbal recall

Verbal recall has been studied in eighteen studies, with varying results on different trimesters or according to parity. For instance, Keenan et al. [22] discovered that from the second to the third trimester, contrasts revealed a significant decline in memory for the pregnant group. For the non-pregnant control group, there were no discernible memory changes. Word list learning tasks have been associated with lower verbal recall in first and third trimesters as reported by [47, 48]. Another study done by de Groot et al., in a comparison to non-pregnant controls (n = 57), pregnant women (n = 71) tested at 14 weeks (i.e. second trimester) of gestation displayed lower immediate and delayed recall on a word-learning task [40]. Another study by Glynn found that after 25 weeks of gestation (second trimester), pregnant women (n = 254) had lower paired-associate learning than non-pregnant controls (n = 48) [38].

Subjective reports suggest that if an antepartum memory deficit exists, it is strongest in the third trimester and resolves shortly after childbirth [13]. Studies using objective measures [21, 35] reported finding impaired verbal recall during pregnancy with lower free recall was most frequently found during the second or third trimester.

According to parity Sharp et al. reported that when compared to non-pregnant controls, multiparous pregnant women remembered fewer words. Primiparous women tended to have worse recall compared to controls [23].

Despite these reports, it’s important to remember that roughly the same number of other studies with comparable sample sizes have also failed to detect any differences in verbal recall performance between pregnant and non-pregnant women [24, 25, 32, 34, 39, 42–45, 49]. Indeed, verbal recall research has produced a variety of results, sometimes even within the same study. For instance, Henry and Sherwin [35] discovered that pregnant women (n = 55) performed worse than non-pregnant controls (n = 21) on verbal paired-associates and word lists’ immediate recall, but they discovered no significant differences in the recall of paragraphs in the same sample of women. Therefore, verbal recall is not affected during pregnancy according to nine high quality studies compared to eight low quality ones.

Verbal recognition

Whereas many studies have included verbal recall tasks, only six have tested recognition. According to Christensen et al. pregnancy-related words were more commonly recognized by women in the third trimester than by those in the second trimester, who, in turn, recognized them more than non-pregnant controls [25]. Wilson et al. on the other hand, discovered that in logical memory task pregnant women had lower story recognition than non-pregnant women [47]. Compared to non-pregnant women, pregnant women performed worse on the story recognition task in the Verbal Paired Associates. The limited amount of evidence indicates that additional research is required. However, these studies collectively suggest that there might be a negligible recognition benefit from pregnancy. On the other hand there were no significant differences concerning recognition in pregnant women compared to non-pregnant controls [2, 23, 35, 50]. Given that the latter four studies are of high quality, their assertion that verbal recognition is unaffected is more credible.

Implicit memory

Implicit memory involves unconscious retrieval of information, such as skills and habits, without explicit awareness. Understanding the potential impact of pregnancy on various faces of memory is of significant interest within the field of psychology and healthcare. Specifically, it is essential to determine whether pregnancy-related changes extend to implicit learning, a crucial aspect of memory that often operates beneath conscious awareness. To address this question, our research yielded nine different studies tackling this topic [22–26, 42, 44, 47, 48] with the exception of two studies worth highlighting [23, 42]. The research conducted by Brindle et al. and Sharp et al. reported findings that diverged from the overall trend observed in the other studies [23]. These two showed that pregnant women performed much worse than non-pregnant controls on Word-Stem Completion Task. It is worth noting that one of these two studies [42] is of low quality making its results less credible. In aggregate, implicit memory is also preserved during pregnancy according to the high-quality studies presented above.

Visuospatial Memory

Visual memory relates to the ability to store and retrieve visual information, such as faces, objects, or spatial arrangements. Researchers have only recently started to take visuospatial or other nonverbal stimuli into account. Studies including visuo-spatial stimuli revealed that pregnancy has not been linked to a decrease in the ability to recall visual stimuli such as faces, objects, or pictures [23, 33, 38, 39, 42, 47, 48]. In one study, it was discovered that pregnant women showed improved face recognition, which was most pronounced for male faces belonging to the same racial group [41].

Verbal- auditory memory

Auditory memory focuses on the retention and recall of auditory information, including sounds, tones, or spoken language. The Rey Auditory Verbal Learning Test (RAVLT) is a neuropsychological tool widely used to assess functions such as attention, memory, and learning ability in the auditory-verbal domain. Four studies [35, 39, 47, 48] have included The Rey Auditory Verbal Learning Test (RAVLT) in their analysis concerning auditory memory. The results of these studies –overall- show that there is actual decline in auditory memory as reported by [35, 47, 48]. Henry and Sherwin [35] reported that compared to never pregnant women, pregnant women recalled fewer words on Trial 1 and at immediate recall at both assessments. In essence, Wilson et al. in both studies [47, 48] reported the same findings. One of the four studies [39], reported no difference in performance between the two groups. Thus, the verbal auditory memory is affected during pregnancy according to three high quality studies.

Prospective memory

Recent research also indicates that prospective memory, a different type of memory, may suffer during pregnancy. It can be characterized as remembering or keeping track of one’s future goals and the ability to remember to perform planned actions or tasks at specific future times, often related to everyday activities and responsibilities. The possibility of future memory deficits is suggested by pregnant women’s self-reports of increased forgetfulness during pregnancy (especially for future events like remembering to run errands) [13]. To identify the cause of the subjective complaints, recent research has started to look at both lab-based and naturalistic measures of prospective memory. Cuttler et al. [39] discovered that pregnant women were less likely than non-pregnant controls to recall the need to call the lab twice (the day before their lab visit and a week later) and to mail an envelope to the lab the day after their lab visit. Pregnant women performed worse than controls on a variety of prospective memory tests, according to two additional studies [37, 45].

The evidence suggests that pregnancy has a small to moderately negative effect on naturalistic prospective memory tasks, which is highlighted by a recent meta-analysis (Anderson & Rutherford, 2012). As a result, prospective memory is actually decreased.

Memory during Postpartum

Memory has been tested in seven studies during both the antepartum and postpartum periods to investigate whether cognitive changes-if present- persist after child birth and early motherhood.

Four out of those seven reported a persistent decline of memory during postpartum. For instance, the pregnant group scored significantly lower than the control group on the Spatial Recognition Memory test at the second trimester and postpartum assessments according to [33]. In the same vein, postpartum women exhibited impairments in working memory, immediate and delayed free recall, prospective memory, and processing speed compared to control women as reported by [50]. Concerning verbal memory specifically, Glynn found that differences in verbal recall memory that emerged during pregnancy persisted when the pregnant group’s performance was compared to that of the non-pregnant group three months after delivery [38]. Another study reported that memory decline during pregnancy persists during postpartum, as memory encoding and retrieval were significantly lower in pregnant women compared to non-pregnant women, and this difference was still present at 32 weeks after delivery [21].

One study explained that postpartum women are more likely to be late with their responses rather than completely missing them, unlike pregnant women [37]. In contrast the remaining two studies [2, 31] showed that the memory decline symptoms were resolved after birth. The overall conclusive result is that memory disturbances and decline persist for weeks or even months after delivery.

Assessment of executive functions and attention during pregnancy

Executive functions, which are primarily regulated by the frontal cortex, encompass a range of cognitive processes. These processes involve attention, planning, the capacity to shift between different ideas (flexibility), the ability to generate novel responses (fluency), problem-solving skills, abstract thinking, and the capability to suppress inappropriate or impulsive reactions.

Attention during pregnancy

Studying the impact of pregnancy on attention is of paramount importance as it provides crucial insights into the multifaceted nature of maternal well-being and child development. Pregnancy is a transformative life stage characterized by hormonal fluctuations, physical changes, and emotional adjustments. Understanding how these physiological and psychological changes influence a pregnant individual’s attentional processes is essential for safeguarding maternal and fetal health.

To gauge attention, a study done by Raz et al. used the Online Continuous Performance Test (OCPT) [46]. As a continuous performance task, the OCPT calls for participants to stay vigilant and respond to the presence or absence of a stimulus. The results of these tests revealed that pregnant women performed worse on attention tasks than controls. In a similar vein [49], reported that when compared to non-pregnant women, pregnant women in their third trimester had significantly lower scores on the attention test D2. Another study found no overall difference between pregnant women and the control group; rather, it identified an in-between group effect among pregnant women only [43]. There was evidence of worse performance in the 3rd trimester of pregnancy on a measure of attentional switching in the TEA Visual Elevator task. Women in their 3rd trimester were slower at attentional switching compared to women in their second trimester and non-pregnant women. In essence, de Groot et al. [27] showed that pregnant women had a significantly smaller pre-cuing benefit at week 36 of pregnancy compared to control women who were not pregnant. This indicates that pregnant women had a loss of selective attention. In contrary, one study done by Roos et al. [51] reported that distressed pregnant women had significantly increased selective attention to fearful faces compared to distressed non-pregnant controls. Another study done by [25] reported that third and second trimester pregnant women performed better on dot probe task when assessing attention. On the other hand [20, 21, 32, 34, 35], found no difference in attention between the two studied groups. No conclusive result can be inferred for this type of cognitive function.

Other executive functions

Intelligence

Only three studies [31, 45, 52], in which none of them found a real effect on pregnancy on intelligence.

Planning

Planning has been studied more frequently in four studies [21, 33, 35, 40, 43]. None of these reported any decline in planning during pregnancy.

Processing speed

Harris et al. and Christensen et al. [2, 34] reported that there was no difference between the studied groups, but Henry and Sherwin’s study [35] found that processing speed had decreased due to pregnancy when utilizing WAIS-III Digit Symbol. Rehbein et al. [52] also reported that there was a significant difference in performance on the TMT between the groups. Specifically, pregnant women showed a slower processing speed on the TMT-A compared to both naturally cycling groups. Thus, there is no conclusive conclusion concerning this type of executive functions.

Cognitive flexibility

Pregnant women showed a slower cognitive flexibility on the TMT-B compared to the low E2 group (non-pregnant women) according to the study of Rehbein et al. [52]. While Harris et al. [34] found no significant difference between the two groups.

Speed of comprehension

Speed and Capacity of Language-Processing Test SCOLP was utilized in one study by Crawley et al. [43] to assess the speed of comprehension in pregnant women. They reported that pregnant women showed worse scores in this task compared to the non-pregnant group and this difference was significant.

Table 4 presents an overall summary of the influence of pregnancy on each category of executive function.

Table 4.

A summary of the influence of pregnancy on each category of executive function

Attention and executive functions	Result
Attention	No conclusive result
Intelligence	No influence of pregnancy on intelligence
Planning	No influence of pregnancy on planning
Processing speed	No conclusive result
Cognitive flexibility	No conclusive result.
Speed of comprehension	Pregnant women showed worse scores compared to the non-pregnant group and this difference was significant

Discussion

The delicate interplay between reproductive physiology and cognitive processes is explored in depth in the complex and multifaceted topic of how pregnancy affects cognitive function. Our study examines cognitive changes that occur during pregnancy through the lenses of different memory types. In our discussion we will explore the potential impact of different predictors, mainly hormonal changes, fragmented sleep, maternal depression, the trade-offs hypothesis, and many others. Our study suggests that pregnancy is linked to a slight decline in cognitive function, which is in such a coherence with other contemporary meta-analyses [8, 54].

Considering the complexity of cognitive function, it is paramount to dissect how pregnancy affects distinct types of memory. The findings suggest that prospective memory is the type of memory that is most negatively impacted during pregnancy. Only when using objective tasks other than the backwards digit span test was it reported that working memory also disrupted. Other types of memory appear to be less significantly impacted by pregnancy. Memory loss in the postpartum period seems to linger in some cases for weeks or even months after giving birth. The other executive functions were also significantly impacted, with speed of comprehension being the most notable. Pregnancy has no impact on intelligence or planning, but we were unable to draw any firm conclusions about other executive functions like cognitive flexibility, attention, or processing speed.

It is crucial to investigate the mechanisms underlying cognitive changes during pregnancy. Both objective and subjective evidence point to impaired cognitive changes during pregnancy, particularly in the area of memory, but the underlying cause has not yet been identified. One of the important questions is whether a change in another physiological process, such as inadequate sleep or hypoglycemia, is a secondary cause of a decline in cognitive function. Pregnancy can have a significant impact on sleep quantity and quality [55], and inadequate sleep can have a negative impact on cognition [56, 57]. According to early self-reporting studies, poor sleep did not appear to be linked to a decline in cognition [58]. There is also no connection between sleep, anxiety, or depression levels and how well pregnant women perform on memory tests, according to research using objective testing measures [22]. In fact, sleep deprivation is a predictor of reported memory loss in pregnant and postpartum women, but it has no effect on actual memory test performance, suggesting that sleep deprivation only affects how forgetful one feels rather than how well one’s memory actually functions [24]. This finding perfectly meshes with Hampson’s [53], where they added another factor which is elevated cortisol- end product of one of the body’s major stress response system- in which neither one has been found related to memory decline. This is consistent with other research showing that sleep deprivation does not cause memory disruption during pregnancy [24, 35, 39, 44].

Another predictor was studied taking into account the number of neurobiological changes that take place throughout a typical pregnancy may also pave the way for affective change. The levels of female reproductive hormones—in particular, estrogen and progesterone— that increase exponentially during pregnancy. These hormones have been found to alter cortical and limbic network activity in a variety of intricate ways [59, 60]. Depression -a very common occurrence during gestational periods- is associated with increased cortisol levels. Chronic stress contributes significantly to depressive symptoms in other, even in non-pregnancy conditions and is linked to increased irritability, low mood, and constant worrying [61]. One study by Marcus et al. [62] found that 20% of pregnant women who were screened on a large scale for depressive symptoms (n = 689) scored in the clinically impaired range. Hampson [53], on the other hand, looked at how those hormones affected affective changes, particularly the emergence of depressive symptoms during pregnancy. Her study was the only one to discover that increased estrogen levels were linked to an increase in working memory errors during cognitive tasks. So to say, estrogen elevation may be a predictor of working memory deterioration.

Many theories have emerged to serve as an explanation to cognitive decline coinciding with gestational periods. One theory is that, in order to meet the high metabolic demands of pregnancy, there is a trade-off that results in cognitive decline during pregnancy [63]. It shows that the limited energy resources are mainly allocated to maintaining fetal growth and preparing the mother for impending motherhood, resulting in a constrained allocation for non-essential cognitive functions like working memory and attention. This lack of energy for cognitive processes leads to detrimental cognitive changes that are primarily brought on by increased neuro-degeneration. However, there is a bright spot in the form of extra energy, which is set aside to support the cognitive adaptations essential for the survival of the mother and her offspring. This extra energy encourages more neurogenesis in particular maternal brain regions, which may help with the cognitive changes required for effective parenting (i.e. emotion recognition, emotion processing and emotion regulation). This last point is emphasized by a study done by Anderson and Rutherford [41] who discovered that pregnant women displayed increased vigilance or recognition of novel male faces, with this tendency being particularly pronounced for faces of men of their own race. They proposed that, given the potential dangers posed by strange males in our ancestors’ evolutionary past, this maternal adaptation may have evolved as a protective mechanism for both mother and child. However, it is unclear whether this kind of adaptive enhancement could account for improved recognition memory for other types of stimuli.

An alternative explanation is that the specific cognitive tasks being considered are not essential for the woman to adjust to her current state, which would explain why cognition appears to be declining. According to this theory, changes in cognition are a result of the brain’s adaptation to motherhood, which reorganizes cognitive functioning to concentrate on abilities required for raising children [54]. However, as suggested by the “cognitive reorganization” theory of pregnancy previously mentioned [54], this may happen at the expense of other cognitive function, such as the memory deficits seen during pregnancy [8]. It’s likely that improved mother-infant relationships and the mother’s capacity to identify her children’s needs are made possible by social cognitive development, which gives offspring a survival advantage.

The maternal brain undergoes significant plasticity during pregnancy as the woman prepares to become a mother and changes her behavioral focus to that of nurturing and caring for her unborn child. According to MRI studies, pregnancy is linked to a decrease in brain volume in humans, which is reversible six months after delivery [9]. This finding emphasizes the idea that the pregnant brain is in a transitional stage and may be undergoing preparation for motherhood. Another study found that gray matter declines in pregnant women, particularly in areas involved in social processes, and that these declines last for at least two years after delivery [12]. According to [12], the decrease in gray matter is thought to be a result of synaptic pruning, which allows the brain to become more specialized for the task of raising children. There is some evidence to support this hypothesis that women have greater social cognition after giving birth, as pregnant women exhibit improved emotion and face recognition [41, 54].

Limitations

Our study faced several limitations, including the inability to conduct a meta-analysis. In this systematic review, the inclusion of 18 cross-sectional studies, 12 cohort studies, and 1 randomized controlled trial presents significant challenges for conducting a meta-analysis. The diversity in study designs leads to substantial variations in methodologies, affecting the comparability of results. Furthermore, each study employs different tools for assessing cognitive function and attention, resulting in inconsistent outcome measures that complicate data aggregation. The heterogeneity is further exacerbated by differences in study contexts and populations. Given these factors, the variability across studies is too extensive to allow for a meaningful quantitative synthesis, rendering a meta-analysis infeasible. Instead, a narrative synthesis was deemed more suitable for summarizing and interpreting the findings across the included studies.

Additionally, the prevalence of cross-sectional studies, with inherent sampling biases and poorly matched control groups, makes it difficult to draw firm conclusions. Publication bias, where studies with significant results are more likely to be published, and language bias, which might have excluded non-English studies, could further skew results. The quality of the included studies varied, potentially affecting the reliability of findings. Search limitations and incomplete reporting in some studies might have led to missing relevant data. Heterogeneity in study populations and outcome measures also posed challenges in comparing and synthesizing results.

Recommendations

It is crucial to address the cognitive changes during pregnancy from multiple angles, including healthcare, workplace policies, legal protections, and public awareness, ultimately supporting pregnant individuals and fostering a more inclusive and understanding society. To comprehensively address cognitive changes during pregnancy, a multi-faceted approach is essential across healthcare, workplace policies, legal protections, public awareness, and lifestyle factors. Healthcare providers should actively engage with pregnant individuals during prenatal care, offering guidance and coping strategies for managing cognitive shifts, supported by accessible mental health services [64]. Public awareness campaigns are crucial to dispel misconceptions surrounding “pregnancy brain” and reduce workplace discrimination [1]. Workplaces should adopt inclusive policies such as flexible work arrangements and reasonable adjustments to support pregnant employees. Legal protections under the American Disabilities Act can safeguard against discrimination based on perceived cognitive changes. Responsible media reporting is necessary to avoid perpetuating stereotypes. Encouraging a healthy lifestyle, including regular exercise and a diet rich in polyphenols, may mitigate cognitive decline [65]. Lastly, future research should prioritize studying attention and executive functions alongside memory to enhance understanding of cognitive changes during pregnancy.

Future studies should employ repeated measures designs whenever possible. Although some studies [48] have included measures of sleep duration and quality, the majority have not. Given the well-known significance of sleep for verbal recall, attention, and working memory performance [66], this is a significant confounding variable. Future research should consider sleep quality because it is frequently disturbed or reduced during pregnancy, especially in late gestation when memory disturbances are most frequently reported, and because it may be even more disturbed in women who are depressed. Finally, surprisingly few studies have examined the negative effects of co-occurring depression and anxiety. Despite compelling evidence that depression can cause the kinds of memory deficits that are believed to occur during pregnancy, the majority of previous studies have used subpar or nonexistent measures of depressive symptoms [67]. Lastly, future research should prioritize studying attention and executive functions alongside memory to enhance understanding of cognitive changes during pregnancy.

Conclusion

In aggregate, our investigation into the complex interaction between pregnancy and cognitive function sheds light on the fascinating dynamics present during this life-changing time for women. Our findings support the notion that pregnancy is indeed linked to a very slight decline in cognitive functioning, with an emphasis on the effect on memory. According to our research, this cognitive shift is most pronounced in verbal recall, with other types of memory and attention showing only minor changes. Despite the abundance of theories and explanations, including those involving hormonal changes, disturbed sleep patterns, and maternal depression, the precise causes of this cognitive decline continue to be a mystery.

This decline in cognitive functioning may raise concerns regarding safety, as impaired memory and attention can affect decision-making and everyday tasks. The investigation into the interplay between pregnancy and cognitive function highlights the need for a multifaceted approach that encompasses healthcare, public awareness, workplace policies, legal protections, responsible media reporting, and promoting healthy lifestyle. By addressing the challenges and complexities of cognitive changes during pregnancy, society can better support and empower pregnant individuals as they navigate this transformative period in their lives. Finally, it is crucial to advocate for continued research in this field to unravel the intricate mechanisms behind cognitive changes during pregnancy and explore potential interventions or strategies to mitigate cognitive decline or help pregnant individuals adapt more effectively. Longitudinal studies tracking cognitive changes across the entire pregnancy journey could yield valuable insights.

Abbreviations

NOS

New Castle Ottawa Scale

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

MINORS

Methodological Index For Non- Randomized Studies

WAIS

Wechsler Adult Intelligence Scale

RAVLT

Rey's Auditory Verbal Learning Test

SILS

Shipley Institute of Living Scale

TEA

Test of Everyday Attention

BADS

Behavioral Assessment of the Dysexecutive Syndrome

WLT

Word Learning Task

SCOLP

Speed and Capacity of Language Processing Test

DKEFS

Delis-Kaplan Executive Function System

WMS

Wechsler Memory Scale

WMS-R

Wechsler Memory Scale-Revised

CST

Concept Shifting Test

LDST

Letter Digit Substitution Test

CANTAB

Cambridge Neuropsychological Test Automated Battery

PASAT

Paced Auditory Serial Addition Test

COWAT

Controlled Oral Word Association Test

TMT

Trail Making Test

OAPAN

Operation Span

OCPT

Online Continuous Performance Test

Authors’ contributions

Conception or design of the work: JY and LAA, Data collection, extraction, and quality assessment JY and MB and AI, supervision, LAA, writing—original draft preparation, MB; Critical revision of the article: LAA and JM ; all authors read and approved the final manuscript.

Funding

This research received no external funding.

Data availability

All data generated or analyzed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

As this is a systematic review, there was no need for Ethics Committee approval.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.