Students’ Emotions Toward Assessments: A Systematic Review

Abstract

This systematic review aims to identify and analyze studies that address emotions related to assessment processes. Using the PRISMA methodology, we reviewed 15 studies published between 2019 and 2024. The search took place between January and April 2025 across the following databases: Scopus, Web of Science, Redalyc, Scielo, ProQuest, Dialnet, and ScienceDirect. The findings show that anxiety emerges as the most frequent emotion, particularly in standardized, oral, or memory-based assessments. In contrast, positive emotions such as hope, pride, and satisfaction appear more prominently in contextualized and collaborative assessments. These results highlight the crucial role of the emotional dimension in assessment processes, as emotions can either enhance or hinder learning outcomes. We conclude that assessment practices require rethinking by integrating the emotional dimension as a formative axis in their design and implementation.

1. Introduction

Emotions play a decisive role in human experience and across multiple domains of life. In education, they are essential to students’ personal, social, and intellectual well-being. Traditionally, research on learning has prioritized cognitive functions while neglecting the emotional dimension. Fernández and Fialho (2016) argue that for many years emotions received little attention because scholars considered them irrational in nature. Bisquerra (2020) notes that a significant shift occurred during the 1990s, when researchers began to recognize the relevance of emotions in education and the need to study them rigorously, along with methods to evaluate them. Hascher (2010) also points out that cognitive sciences largely ignored the role of emotions in learning processes. Today, scholars widely acknowledge that emotions directly influence learning, motivation, academic achievement, and students’ holistic development. Indeed, the growing body of research on emotions (Hargreaves 1998; LeDoux 1999; Shapiro 2010) demonstrates that cognition and emotion, viewed as complementary, offer deeper insight into educational phenomena, particularly those related to learning and knowledge construction.

Researchers define emotions as intense, short-term reactions to unexpected stimuli that stem from biological needs (Bisquerra 2001; Bisquerra and López-Cassà 2021). These complex phenomena affect consciousness and school learning (Palmer 2017; Pekrun et al. 2002; Pekrun 2023; Pekrun and Perry 2014; Mellado et al. 2014). They involve coordinated processes across psychological subsystems, including physiological, cognitive, and motivational aspects (Pekrun 2006, 2024). Emotions form an intricate network that connects individuals to their environment and plays a fundamental role in human behavior (Le Breton 2013; Segura et al. 2020).

Emotions arise from the information we process in our surroundings (Bisquerra 2000; López-Cassá and Bisquerra 2023). Their intensity depends on subjective evaluations shaped by prior knowledge and beliefs. In other words, emotions emerge from what individuals value as important. They do not only respond to present stimuli but can also be triggered by past memories or future anticipations (Damasio 2010).

Multiple theoretical perspectives have contributed to the understanding of emotions: psychophysiological (Pekrun 2006, 2024; Le Breton 2013; Segura et al. 2020), evolutionary (Darwin 1872), neuroscientific (LeDoux 1999), and integrative (Capafons and Dolores 1984; Miguel Tobal and Cano Vindel 1990). In education, achievement emotions represent a particularly relevant category. Schutz and Pekrun (2007) describe these as complex, involving affective, cognitive, motivational, and expressive components. Emotions such as enjoyment, hope, pride, anxiety, shame, and boredom strongly influence students’ motivation, learning, academic performance, and identity development (Pekrun et al. 2011).

Negative emotions like anxiety can impair academic performance because they consume essential cognitive resources (Pekrun 2006, 2024). Conversely, positive emotions such as hope and resilience foster motivation and academic success (Pekrun et al. 2011). Méndez López and Peña Aguilar (2013) even argue that emotions can determine whether a student persists with or abandons a learning task. From this perspective, emotions act as powerful forces that either facilitate knowledge acquisition or create barriers to it.

In schools, students experience a wide range of positive and negative emotions. Positive emotions strengthen motivation, engagement, and achievement, while negative emotions often hinder learning and personal growth (Cumbes-Chávez 2018). The mobilization of emotions in school settings does not occur in isolation: teachers, families, and institutions all shape students’ emotional experiences. Academic factors such as teaching methods and assessment processes also directly influence the emotional climate, affecting learning, social interaction, and well-being.

Emotions not only shape learning in general, but also play a critical role in assessment. Fear, anxiety, and distress represent natural responses to challenging situations, especially exams (Sierra et al. 2003). As Gusils et al. (2021) emphasize, “taking an exam is one of the most stressful situations in the academic life of students” (p. 33). Test anxiety—a common phenomenon among students—manifests as excessive worry, distress, and anticipation of failure in evaluative contexts (Putwain and Aveyard 2018). This form of anxiety negatively affects academic performance, self-esteem, and interest in school activities (Furlan 2006).

Assessments, also referred to as tests or examinations, represent essential academic processes designed to verify students’ knowledge, skills, and abilities (Bueno 2021; Zambrano-Vélez et al. 2023). These evaluations often elicit a variety of emotions. Typically based on uniform curricula and standardized grading methods, they may overlook the richness of cultural diversity and the different socioeconomic backgrounds of students. Such disconnection can foster feelings of exclusion and discouragement, which negatively affect both students’ emotional states and the overall school climate (Pulido Acosta and Herrera Clavero 2017).

Several studies highlight the emotional effects of examinations and their relationship with either academic success or failure (Gusils et al. 2021; Gutiérrez-Vergara et al. 2020; Obregón-Cuesta et al. 2022; Paoloni and Vaja 2013; Pekrun et al. 2002; Pekrun 2023). Students’ emotions significantly influence both short- and long-term development in learning, interest, and knowledge acquisition (Domínguez Lara 2018). These experiences encompass a wide spectrum of positive and negative emotions, depending on situational and contextual factors. Although evidence exists, researchers emphasize the need for further inquiry into students’ emotional experiences in academic settings (Pekrun et al. 2002; Pekrun 2023).

In this review, we focus on the recent 2019–2024 corpus to clarify three complementary dimensions that are pivotal for both scholarship and practice: who currently shapes the field, where research is being produced and in which educational settings, and how emotions and assessment are conceptualized and operationalized. Establishing the intellectual configuration of the field helps locate the theoretical anchors and avoid citation biases; mapping the geographical and institutional settings speaks to external validity, equity, and transferability across systems; and examining conceptual definitions, research designs, and measurement instruments determines the comparability and credibility of findings.

The intentions formulated in the preceding paragraph underscore the relevance of examining how contemporary research addresses students’ emotions in assessment contexts. These intentions are channeled through the following overarching question, which guides and structures the synthesis:

What is the current state of academic production published between 2019 and 2024 on students’ emotions toward assessments, considering theoretical perspectives, methodological approaches, and key findings?

2. Materials and Methods

We followed the PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) to ensure transparency, completeness, and replicability in this systematic review (Moher et al. 2009). We designed the search strategy across databases and academic sources related to psychology and the social sciences, including Scopus, Web of Science, Redalyc, Scielo, ProQuest, Dialnet, and ScienceDirect. We applied descriptors and Boolean operators in both English and Spanish. The search targeted article titles to achieve a precise thematic delimitation and to optimize the relevance of the selected studies (see

Table 1. Search Equations.

Language	Databases	Search Equations
English	Scopus, Web of Science, ProQuest, ScienceDirect	(emotions OR “achievement emotions”) AND (assessment OR evaluation OR test) AND student
Spanish	Dialnet, Redalyc, Scielo	emociones AND estudiantes AND exámenes

). No backward (reference-list) or forward (citation-index) searching was conducted, nor were gray-literature sources queried; all records were retrieved from the databases listed. The descriptors used were the following:

Inclusion and Exclusion Criteria

We included peer-reviewed articles published 2019–2024, in English or Spanish, with full text available, and with the co-occurrence of emotions and assessment in the title to ensure a focus on educational contexts. We excluded studies with restricted access, those that did not explicitly link emotions and assessment in education, and those with conceptual limitations—(i) no explicit link between emotions and assessment; (ii) substitution of adjacent constructs without operationalizing emotions or validated measures; (iii) non-educational contexts or without assessment processes (e.g., clinical or occupational assessments) (see Figure 1).

Guiding Questions

This review relied on a set of guiding questions that structured both the quantitative and qualitative synthesis of the selected studies. The systematic review addressed the following research questions:

Q1. Who are the most influential and frequently cited authors and researchers in the literature on emotions and assessment during 2019–2024?

Q2. In which countries were the studies on emotions related to assessment primarily conducted between 2019 and 2024, and what educational contexts did they examine?

Q3. How do the studies published during 2019–2024 conceptualize emotions and assessment?

Q4. What are the main research designs, methodological approaches, and instruments employed?

Q5. What forms of assessment dominate the recent literature, and what emotional patterns among students are associated with them according to the methodological designs and approaches analyzed?

Q6. What key findings, main conclusions, and convergent results have been reported in studies conducted between 2019 and 2024 regarding the emotional impact of assessment on students?

The next section presents a synthesis of the results obtained from the literature search.

3. Results

3.1. Most Relevant Authors and Researchers

Among the reviewed studies, Reinhard Pekrun stands out as the most frequently cited researcher (see

Table 2. Most Frequently Cited Authors. Time Range denotes the earliest and latest publication years of works by each author cited within the included studies; it is not the review’s inclusion window.

Author(s)	Number of References	Time Range	Key Themes
Reinhard Pekrun et al.	25	2000–2019	Academic emotions: anxiety, motivation, control-value theory
José De la Fuente et al.	16	2015–2020	Self-regulation and coping: academic stress, academic performance
Patrick R. Goetz et al.	12	2002–2019	Classroom emotions: performance, enjoyment, boredom
Laura Furlan et al.	10	2006–2023	Test anxiety: perfectionism, psychological interventions
Peter D. MacIntyre et al.	6	1989–1999	Foreign language learning: anxiety, motivation, communication
Natalio Extremera et al.	5	2004–2013	Emotional intelligence: burnout, engagement, emotional well-being
Elaine K. Horwitz et al.	5	1986–2010	Foreign language anxiety: measurement scales, learning outcomes
Pablo Fernández-Berrocal et al.	5	1999–2008	Emotional intelligence in education: emotional education, instrument validation
Fernando G. Arana et al.	5	2002–2016	Assessment-related emotions: anxiety, evaluation, motivation, perfectionism
Samuel A. Domínguez-Lara et al.	5	2016–2018	Psychometric validation: anxiety, emotional regulation
John D. Mayer and Peter Salovey	4	1995–2008	Emotional intelligence theory: emotion and education
Sükran Aydın et al.	4	2006–2020	Foreign language anxiety: teacher feedback, individual differences

). His 25 contributions focus on the development of the control-value theory and the analysis of emotions such as anxiety, boredom, and enjoyment in academic contexts. José De la Fuente follows with 16 references, emphasizing self-regulation, coping with academic stress, and student performance from a psychoeducational perspective. Patrick R. Goetz also emerges with 12 citations dedicated to the study of classroom emotions. In Latin America, Laura Furlan holds the strongest presence with 10 citations, mainly addressing test anxiety, perfectionism, and psychological interventions.

Additional significant contributions come from Peter D. MacIntyre (6 references), who explores foreign language anxiety and communicative motivation, and Elaine K. Horwitz (5 references), known for creating the widely used Foreign Language Classroom Anxiety Scale (FLCAS). Natalio Extremera and Pablo Fernández-Berrocal, with 5 references each, have played a key role in validating the Trait Meta-Mood Scale (TMMS) and examining emotional well-being in education. Fernando G. Arana and Samuel A. Domínguez-Lara, both with 5 citations, contribute to psychometric evaluation and emotional regulation research. Finally, John D. Mayer and Peter Salovey appear with 4 references each, providing the theoretical foundation of emotional intelligence as an ability, which has shaped subsequent studies on emotions and academic performance.

Together, these authors have shaped an interdisciplinary field characterized by consistent theoretical frameworks and rigorous methodologies, offering valuable insights into the complex relationship between emotions and assessment in diverse educational contexts.

3.2. Q2: Geographical Distribution

The review reveals a strong concentration of studies in Latin America (8 studies) with a focus on empirical research adapted to the educational contexts of countries such as Colombia, Ecuador, Argentina, Mexico, and Chile (see Figure 2). In Colombia, Abad et al. (2021), Ávila-Toscano et al. (2021), and Riaño-Rodríguez (2024) examine test anxiety from different perspectives: teacher–student perceptions, emotional and cognitive factors, and technology-mediated pedagogical interventions, respectively. In Ecuador, García-Beracierto et al. (2024) and Zambrano-Vélez et al. (2023) emphasize the influence of anxiety and other emotions on academic performance, highlighting the need for emotional regulation strategies. In Argentina, Furlan and Martínez-Santos (2023) assess the effectiveness of a cognitive–behavioral intervention designed to reduce test anxiety, while in Mexico, González-Peralta and Sánchez-Aguilar (2023) analyze emotions triggered by oral mathematics assessments. From Chile, Gutiérrez-Vergara et al. (2020) investigates academic emotions among kinesiology students at different evaluative stages.

In Europe, 4 studies originate mainly from Spain and Turkey. In Spain, Bueno (2021) links stress to memory and proposes assessment models that foster learning; De la Fuente et al. (2020) explore the role of self-regulation and situational stress in shaping academic emotions; and Obregón-Cuesta et al. (2022) validate a psychometric scale for detecting negative emotions associated with poor academic performance. In Turkey, Denkci Akkaş et al. (2020) present a theoretical framework on test anxiety in foreign language examinations and recommend specific teacher training as a critical measure to reduce its effects.

In Asia, Oceania, and North America, studies from China, New Zealand, and Canada focus on the impact of digital technologies on students’ emotional experiences. Gao (2024) analyzes the role of artificial intelligence as an emotional mediator; Riegel and Evans (2021) highlight the emotional benefits of online assessment; and Harley et al. (2021) examine negative emotions in computer-based examinations, considering both personal and contextual factors.

The following figure summarizes the most relevant studies.

3.3. Q3: Emotions and Assessment

The reviewed studies conceptualize emotions as complex psychological processes closely linked to academic performance, self-regulation, and decision-making in evaluative contexts. From a functional perspective, Riaño-Rodríguez (2024) defines emotions as internal experiences that students must recognize, confront, and regulate, transforming them into tools that provide greater confidence when facing demanding academic situations such as examinations.

Across the studies, anxiety consistently emerges as a specific type of academic emotion. García-Beracierto et al. (2024) describe anxiety as an adaptive alert response to perceived threats that manifests at physical, cognitive, and behavioral levels. Building on this framework, Furlan and Martínez-Santos (2023) characterize test anxiety as intense worry about failure, diminished self-esteem, fear of social evaluation, and physiological symptoms such as stuttering or active avoidance of tests. Similarly, Denkci Akkaş et al. (2020) propose a more nuanced classification, distinguishing among state, trait, and situational anxiety, all of which manifest through physiological and cognitive symptoms under conditions of uncertainty.

From a neuropsychological and psychoeducational perspective, Bueno (2021) situates stress within the broader field of emotions, defining it as a physiological and emotional reaction to stimuli perceived as threatening. He categorizes stress into positive, tolerable, and toxic levels, depending on intensity and the availability of support. Complementing this view, Abad et al. (2021) associate evaluative anxiety with negative emotions such as frustration, insecurity, fear, tension, and unease, all of which can disrupt academic behavior.

Adopting a broader theoretical lens, Gao (2024) defines academic emotions as “students’ emotional experiences related to academic processes such as teaching and learning, including enjoyment, hopelessness, boredom, anxiety, anger, and pride” (p. 1541). This definition aligns with Pekrun’s control-value theory (Pekrun 2024), which posits that academic emotions depend on perceived control over a task and its subjective importance. The theory classifies emotions across three dimensions: valence (positive or negative), activation (high or low), and focus (on the activity or the outcome). This framework explains how emotions such as enjoyment, hope, and anxiety influence motivation and performance.

The reviewed studies also provide a broad, multifactorial view of assessment, framing it both as a technical process of measurement and as a significant event with deep cognitive, emotional, and social implications for students. Riaño-Rodríguez (2024) conceptualizes assessment as a standardized event that generates emotional and cognitive pressure, directly shaping students’ life projects. From a neuroeducational perspective, Bueno (2021) describes assessment as an educational tool that not only grades but also guides what students learn, how they learn, and how they consolidate—or erode—their memory.

From a clinical approach, Furlan and Martínez-Santos (2023) view assessment as an achievement situation that triggers self-evaluation of abilities and may provoke anxiety, particularly in oral contexts, due to their public and socially exposed nature. This perspective aligns with Denkci Akkaş et al. (2020), who argue that academic assessments represent performance situations in which perceived social or personal consequences intensify emotional responses.

In foreign language learning contexts, Abad et al. (2021) emphasize that oral tests assess both linguistic mastery and affective factors that directly shape performance. Complementing this perspective, Gao (2024) cites Basaknezhad et al. (2013), who describe assessment as a situation that can undermine emotional security and weaken students’ coping capacities.

Finally, Harley et al. (2021) argues that assessments constitute achievement situations with high emotional demands, capable of evoking anxiety, shame, or hopelessness depending on the type of test and its perceived relevance. Several studies support this position, noting that assessment format (oral, written, standardized) and evaluative climate (pressure, feedback, consequences) determine the intensity and direction of students’ emotional responses (Abad et al. 2021; Ávila-Toscano et al. 2021; Bueno 2021; De la Fuente et al. 2020; Denkci Akkaş et al. 2020; Gao 2024; Obregón-Cuesta et al. 2022; Zambrano-Vélez et al. 2023).

3.4. Q4: Methodological Approaches

The reviewed studies reveal a predominance of quantitative approaches (11 studies), followed by qualitative (3) and mixed-methods designs (1). This predominance reflects an underlying epistemology that privileges standardized measurement, replicability, and statistical generalization as criteria of scientific legitimacy (see

Table 3. Methodological characteristics of the reviewed studies.

Author	Approach/Design	Instruments	Type of Analysis
García-Beracierto et al. (2024)	Quantitative Descriptive	Questionnaire	Descriptive analysis
Zambrano-Vélez et al. (2023)	Quantitative Descriptive	Questionnaire	Descriptive analysis
González-Peralta and Sánchez-Aguilar (2023)	Quantitative Descriptive	Questionnaires	Descriptive and correlational analysis
De la Fuente et al. (2020)	Quantitative Correlational	Questionnaires	Multivariate analysis (SEM, ANOVA, MANOVA)
Gutiérrez-Vergara et al. (2020)	Quantitative Correlational	Questionnaire	Inferential analysis (t-tests)
Abad et al. (2021)	Quantitative Correlational	Questionnaires	Correlational and regression analysis
Obregón-Cuesta et al. (2022)	Quantitative Correlational	Questionnaires	Multivariate analysis
Harley et al. (2021)	Quantitative Correlational	Questionnaires	Multilevel statistical modeling
Riegel and Evans (2021)	Quantitative Correlational	Questionnaires	Inferential analysis (ANOVA, regression)
Gao (2024)	Quantitative Quasi-experimental	Scales	Inferential analysis (Student’s t-test)
Ávila-Toscano et al. (2021)	Quantitative Correlational	Scales	Correlational and predictive analysis
Riaño-Rodríguez (2024)	Qualitative Descriptive-Interpretative	Interviews, focus groups, observations	Qualitative coding and triangulation
Bueno (2021)	Qualitative Theoretical Review	Scientific documents	Argumentative analysis and narrative synthesis
Denkci Akkaş et al. (2020)	Qualitative Conceptual Review	Documentary review	Theoretical-interpretative analysis
Furlan and Martínez-Santos (2023)	Mixed Quasi-experimental with case study	Questionnaires and interviews	Mixed analysis (qualitative clinical and quantitative)

). Within this framework, researchers emphasize the need to quantify the influence of emotions on academic performance, allowing for analyses and targeted contributions that clarify the role of emotions in assessment processes and their impact on educational quality.

The following table consolidates the methodological characteristics of all reviewed studies.

Correlational designs appear most frequently (7 studies), focusing on relationships between affective and cognitive variables of assessment. For example, Harley et al. (2021) employed multilevel analysis to explore links between academic emotions and performance in digital environments. Abad et al. (2021) and Ávila-Toscano et al. (2021) examined correlations among anxiety, coping, engagement, and emotional intelligence. Although these designs identify patterns of co-occurrence, their inability to establish causality limits explanatory power.

Descriptive designs (3 studies) also appear in the literature. García-Beracierto et al. (2024) measured emotions such as enjoyment, hope, pride, anger, anxiety, shame, hopelessness, and relief at three stages (before, during, and after assessment) while considering demographic factors such as gender and age. Zambrano-Vélez et al. (2023) described emotional states before and during assessment, identifying anxiety and hope as predominant prior to evaluation, while emotions such as enjoyment and pride varied during the process. González-Peralta and Sánchez-Aguilar (2023) coded students’ reported emotions (valence and intensity) before, during, and after an oral practice exam. These studies fulfilled an exploratory function by characterizing emotional presence, frequency, and intensity, highlighting anxiety, fear, and frustration across evaluative stages. Structured questionnaires and Likert-type scales predominated, although these instruments did not capture the contextual or experiential conditions shaping emotions.

One quasi-experimental study stands out for its rigor. Gao (2024) conducted a pretest–posttest design with a control group to evaluate the effects of targeted interventions on test anxiety using standardized measures such as the Westside Test Anxiety Scale and the AEQ. This design represents methodological progress by moving closer to causal inference, although such approaches remain marginal in the field. A notable gap persists in research on how artificial intelligence–driven educational applications influence academic emotions and test anxiety among university students.

Regarding instruments, researchers employed eight questionnaires and three psychometric scales, including the AEQ, TMMS-24, and UWES-S. Their use reflects an interest in capturing affective constructs through observable variables, which facilitate statistical treatment. Studies by De la Fuente et al. (2020) and Obregón-Cuesta et al. (2022) stand out for applying multivariate models (SEM, ANOVA, MANOVA) to analyze complex relationships among emotions, self-regulation, stress, and motivation.

From a critical standpoint, the methodological dominance of quantitative–correlational designs indicates that the field still prioritizes measurability over situated understanding of emotional phenomena. Although researchers recognize emotions as relevant variables in assessment, their treatment often remains anchored in linear, controlled models that reduce students’ subjectivity to numerical scores while neglecting historical, social, and relational contexts.

By contrast, the few qualitative and mixed-methods studies provide richer perspectives. Riaño-Rodríguez (2024) used interviews, observations, and interpretive coding to generate a holistic account of emotions in interaction with pedagogical practices. Similarly, Furlan and Martínez-Santos (2023) combined quantitative scales with semi-structured interviews, integrating measurable outcomes with students’ lived experiences in assessment contexts.

The lack of longitudinal and rigorous experimental studies limits understanding of how evaluative emotions evolve over time and how cumulative effects of anxiety, self-efficacy, or academic stress shape students’ trajectories. Although the prevalence of cross-sectional studies provides useful diagnostic snapshots, it cannot capture the dynamic and progressive ways in which emotions influence educational pathways.

3.5. Q5: Emotional Patterns by Type of Assessment

The reviewed studies clearly demonstrate that students experience differentiated emotional patterns depending on the type, format, and context of the assessment. Emotions vary significantly according to perceived difficulty, prior emotional preparation, and exam modality (oral, written, standardized, practical, etc.). One of the most common patterns involves negative anticipatory emotions such as anxiety, fear, frustration, shame, uncertainty, and hopelessness, particularly in high-stakes standardized tests. Riaño-Rodríguez (2024) highlights that the lack of emotional preparation exacerbates these states, undermining both mental health and academic performance. Bueno (2021) distinguishes between assessment formats, showing that closed, memory-based exams intensify anxiety, whereas practical or oral evaluations redirect stress into cognitive functions that support learning.

In university contexts, García-Beracierto et al. (2024) identify somatic symptoms—including insomnia, tension, and obsessive thoughts—associated with anxiety before cumulative evaluations. Furlan and Martínez-Santos (2023) also describe anticipatory anxiety, shame, avoidance of evaluators, and low self-efficacy, especially in oral assessments. In foreign language contexts, Denkci Akkaş et al. (2020) report intrusive thoughts, loss of confidence, and physiological symptoms such as sweating or muscle rigidity, noting that perceived social judgment intensifies emotional vulnerability.

Abad et al. (2021) document an emotional pattern tied to environmental perception, where students focus their anxiety on visible mistakes or external distractions—factors often minimized by students but magnified by external observers. Gao (2024) reports a notable shift: students who initially expressed negative emotions such as shame and hopelessness evolved toward pride, hope, and enjoyment after a targeted educational intervention, illustrating the potential for emotional transformation through appropriate didactic strategies. From an experimental perspective, Harley et al. (2021) show that students with strong negative emotional profiles (particularly women) achieve lower performance. They also demonstrate that computerized assessments reduce anxiety compared to paper-based formats, fostering greater emotional regulation and enjoyment.

The reviewed literature describes a wide variety of assessment types (see

Table 4. Types of Assessment, Predominant Emotional Patterns, and Representative.

Type of Assessment	Number of Studies	Predominant Emotional Patterns	Representative Authors
Traditional written assessment	5	Anticipatory anxiety, tension, frustration, negative automatic thoughts	Bueno (2021); Harley et al. (2021); De la Fuente et al. (2020); Gutiérrez-Vergara et al. (2020); Zambrano-Vélez et al. (2023)
Individual oral assessment	4	Shame, avoidance, sweating, insecurity, cognitive blocking	Furlan and Martínez-Santos (2023); Abad et al. (2021); Denkci Akkaş et al. (2020); Gao (2024)
Foreign language assessment	3	Fear of social judgment, language anxiety, muscle rigidity, communicative withdrawal	Denkci Akkaş et al. (2020); Abad et al. (2021); Gao (2024)
Digital/computer-based assessment	2	Emotional regulation, reduced anxiety, greater control and enjoyment	Harley et al. (2021); Gao (2024)
High-stakes standardized assessment	2	Pressure, hopelessness, severe anxiety, intrusive thoughts	Riaño-Rodríguez (2024); Bueno (2021)
Formative/low-stakes assessment	1	Emotional security, intrinsic motivation, reduced emotional tension	Gao (2024)

), whose emotional effects differ considerably depending on format, demands, educational context, and students’ emotional preparation. Six broad categories emerge: standardized tests, traditional written and oral assessments, foreign language evaluations, digital assessments, formative assessments, and low-stakes evaluations. This classification builds on the contributions of Riaño-Rodríguez (2024), Bueno (2021), Abad et al. (2021), Denkci Akkaş et al. (2020), Harley et al. (2021), and Gao (2024). Among these, oral assessments in foreign languages stand out. Abad et al. (2021) and Denkci Akkaş et al. (2020) emphasize that anxiety undermines students’ oral performance, particularly through communicative apprehension—a form of shyness characterized by fear or anxiety when speaking.

The following section presents a quantitative synthesis.

Overall, the identified emotional patterns can be grouped according to the stages of the assessment cycle. Before the assessment, anticipatory anxiety, insomnia, and ruminative thoughts dominate, as documented by García-Beracierto et al. (2024) and Gutiérrez-Vergara et al. (2020) in their studies on pre-exam symptomatology in cumulative evaluations. During the assessment, emotions such as shame, cognitive blocking, physiological tension, and avoidance emerge. These phenomena are extensively described by Furlan and Martínez-Santos (2023) and Denkci Akkaş et al. (2020) in oral and foreign language assessment contexts. After the assessment, emotional responses range from relief and pride to hopelessness, depending on perceived outcomes. Gao (2024) highlights this variability, noting positive emotional evolution following an educational intervention.

From a didactic and pedagogical perspective, these findings underscore the need to rethink the design and function of assessments by integrating the emotional dimension as a criterion of formative quality. The centrality of emotions such as anxiety, shame, and fear in high-stakes evaluations suggests that emotional suffering does not occur as a collateral effect but rather as a structural component of current assessment systems. Recognizing this reality requires a transformation of the evaluative paradigm—from a logic of selection and control to one of support, understanding, and holistic student development.

3.6. Q6: Results and Conclusions of the Reviewed Studies

The review shows that academic assessment processes constitute emotionally significant experiences for students, particularly in highly demanding institutional contexts or in environments lacking emotional support. Anxiety, anticipatory stress, and feelings of inadequacy appear frequently, especially among first-year students and in high-demand programs such as health sciences and education (García-Beracierto et al. 2024; Ávila-Toscano et al. 2021).

The type and format of assessment act as modulators of emotional impact. Memory-based or standardized evaluations generate higher levels of emotional tension and cognitive blocking (Riaño-Rodríguez 2024; Bueno 2021), whereas applied, contextualized, and emotionally mediated formats foster more positive and functional affective responses that support learning. Technology-mediated platforms, when guided pedagogically, also enhance self-efficacy and reduce evaluation-related anxiety (Riaño-Rodríguez 2024).

The studies further reveal key differences depending on the specific type of assessment. In oral examinations in foreign languages, students often report emotions such as shame, fear of social judgment, and linguistic withdrawal, triggered by public exposure and perceived evaluation of communicative competence (Denkci Akkaş et al. 2020; Abad et al. 2021). In contrast, well-designed formative assessments show a positive emotional shift, with students moving from anxiety to pride and satisfaction (Zambrano-Vélez et al. 2023).

Several pedagogical and psychoeducational interventions proved effective in transforming emotional states related to assessment. Furlan and Martínez-Santos (2023) demonstrated that cognitive–behavioral interventions reduce procrastination, increase self-confidence, and improve performance in oral contexts. Complementary findings from De la Fuente et al. (2020) and Harley et al. (2021) show that academic emotions such as anxiety, pride, and enjoyment significantly mediate the relationship between assessment, performance, and self-regulated learning.

Taken together, these results support the claim that assessment should not be understood solely as a technical measurement procedure but as a formative event with high emotional weight. Assessments designed without considering the affective dimension often produce negative effects on student well-being. Conversely, when intentionally incorporating emotional mediation, assessments can serve as pedagogical tools that foster socioemotional development, motivation, and improved academic performance.

Limitations

A major limitation is the predominance of cross-sectional designs (Ávila-Toscano et al. 2021; Gutiérrez-Vergara et al. 2020), (see

Table 5. Limitations of the included studies.

Article	Limitation	Impact on Generalization and Comparability
Ávila-Toscano et al. (2021)	Design: cross-sectional; convenience sampling	Does not allow causal inference, limits generalization, and the explanatory power is low (3–5%).
Gutiérrez-Vergara et al. (2020)	Design: observational, cross-sectional	Does not permit causal inferences; experimental designs are needed to validate relationships.
Harley et al. (2021)	Design: not an RCT (Randomized Controlled Trial)	Potential confounders cannot be ruled out, so firm causal conclusions are not supported.
Gao (2024)	Design: non-experimental; convenience sampling	Does not allow attributing causal effects of AI/ChatGPT on emotions or anxiety.
Furlan and Martínez-Santos (2023)	Design: single-case (N = 1); quasi-experimental (pre–post)	Findings are not generalizable and temporal stability is not established.
Abad et al. (2021)	Sample and external validity: small case sample in a specific program	Findings are contextual and difficult to extrapolate to other subjects or institutions.
De la Fuente et al. (2020)	Sample and external validity: only Spanish universities	Lack of cross-cultural contrast limits generalization and intercultural invariance.
Harley et al. (2021)	Sample and external validity: gender imbalance (low n of men)	Gender comparisons require caution and have limited generalizability.
Gutiérrez-Vergara et al. (2020)	Sample and external validity: convenience sampling in one degree program and university	Transferability to other programs and institutions is reduced.
González-Peralta and Sánchez-Aguilar (2023)	Sample and external validity: mostly male students	Results cannot be confidently extrapolated to female students.
Obregón-Cuesta et al. (2022)	Sample and external validity: reported difficulty generalizing	Population and ecological external validity of the instrument is compromised.
Riegel and Evans (2021)	Sample and external validity: convenience sampling with low attendance (~25%)	The sample may be unrepresentative, introducing selection bias.
Ávila-Toscano et al. (2021)	Instrumentation and data collection: self-reports; low situational specificity	Lack of situational precision hinders replication and cross-context comparisons.
Furlan and Martínez-Santos (2023)	Instrumentation and data collection: self-reports/interviews	Recall bias may be present; observational data are needed to corroborate findings.
Gutiérrez-Vergara et al. (2020)	Instrumentation and data collection: long questionnaire (69 items)	Length-related fatigue may bias responses and affect measurement validity.
Harley et al. (2021)	Instrumentation and data collection: low internal consistency for anger subscales (discarded)	Discarding anger subscales leaves the negative emotional spectrum undercovered and limits comparability.
Riegel and Evans (2021)	Instrumentation and data collection: questionable consistency (shame/anger); no temporal phases	Internal validity is affected and the temporal dynamics proposed by CVT (Control–Value Theory) are not examined.
González-Peralta and Sánchez-Aguilar (2023)	Instrumentation and data collection: teacher administered the oral test and the questionnaire	The dual teacher–evaluator role may bias students’ responses.
Denkci Akkaş et al. (2020)	Instrumentation and data collection: lack of consensus and adequate tools	The lack of consensus and robust instruments reduces comparability and the strength of findings.

for a study-level summary of limitations) which preclude robust causal inferences and obscure how evaluative emotions evolve over time. Although some studies used predictive analyses or multilevel/complex models (Harley et al. 2021), the absence of longitudinal designs or randomized controlled trials (RCTs) limits attributing emotional changes to specific variables or interventions. Likewise, single-case designs (Furlan and Martínez-Santos 2023) restrict generalizability to broader educational populations.

Sample and external validity. Several investigations relied on samples confined to specific geographic, institutional, or disciplinary contexts—e.g., Spanish universities or kinesiology students—which compromises external validity (De la Fuente et al. 2020; Obregón-Cuesta et al. 2022). In addition, intentional or convenience sampling increases selection-bias risk, affecting representativeness (Gutiérrez-Vergara et al. 2020; Riegel and Evans 2021).

Instrumentation and data collection. The reliance on self-reports, questionnaires, and interviews is another challenge, as these methods depend on memory and verbal self-regulation and may introduce cognitive or social-desirability bias (Ávila-Toscano et al. 2021; Gutiérrez-Vergara et al. 2020). Some studies also report length/complexity of instruments that may hinder completion and compromise response quality (Furlan and Martínez-Santos 2023). In some cases, predictive models did not clearly identify the conditions under which factors such as perfectionism or procrastination affect evaluative anxiety (Ávila-Toscano et al. 2021). Moreover, two-time-point assessments (pre/post) fail to capture longer-term emotional trajectories (Furlan and Martínez-Santos 2023).

4. Discussion

This analysis demonstrates that emotions form a constitutive, rather than external, component of the assessment process, underscoring the “emotional turn” in educational sciences (Bisquerra 2020; Fernández and Fialho 2016). The results consistently show that academic evaluation activates a complex range of emotions that directly influence learning, performance, and students’ subjective experience. According to Puente-Díaz and Puerta-Sierra (2025), both cognitive processes and the emotional responses that accompany them are essential for evaluating, constructing, and validating creative ideas in innovative educational settings. This emotional weight is particularly salient in high-demand academic environments or in settings with insufficient emotional support, where feelings of anxiety and inadequacy become pervasive, especially among first-year students (García-Beracierto et al. 2024; Ávila-Toscano et al. 2021).

From an epistemological standpoint, the findings challenge the traditional dichotomy between cognition and affect. Emotions, understood as neurobiological, social, and cultural processes, shape the meaning of assessment and condition not only how students approach examinations but also how they construct their identity as learners. As Pekrun (2006, 2024) and LeDoux (1999) argue, emotional activation significantly impacts attentional and memory resources, thereby determining academic performance. Despite growing scientific consensus, many educational systems continue to operate under outdated assessment models that ignore emotional processes. This disconnection is especially visible in Latin American contexts, where structural inertia limits innovation and perpetuates emotionally harmful evaluative practices.

One of the most relevant contributions of this review is the consolidation of anxiety as the predominant emotion in evaluative contexts, whether anticipatory, situational, or trait-based. Chang (2025) shows that students’ belief systems regarding academic self-concept mediate the relationship between perfectionism—conceived as an intrinsic evaluative philosophy—and emotions, with anxiety emerging as the most disruptive and mobilizing factor within this self-demanding logic. The prevalence of anxiety is supported by both quantitative empirical studies (Abad et al. 2021; Harley et al. 2021) and clinical or qualitative approaches (Furlan and Martínez-Santos 2023), validating its transdisciplinary relevance. The identification of anxiety as the dominant academic emotion is not a neutral finding; rather, it signals a systemic issue in how institutional environments normalize emotional distress as a byproduct of academic achievement. This normalization raises ethical concerns and calls for a reconfiguration of assessment that prioritizes emotional well-being alongside academic rigor. Negative emotions tend to dominate in standardized, oral, and memory-based written assessments, whereas contextualized, collaborative, or technologically mediated formats more often elicit positive emotions such as pride, enjoyment, and hope (Gao 2024; Zambrano-Vélez et al. 2023). These patterns highlight the structural role of assessment design in shaping students’ emotional experiences, suggesting that changes in evaluation formats can directly influence motivation and engagement.

In particular, well-designed formative assessments (such as those incorporating peer feedback, reflective journaling, and self-assessment tools) have been shown to reduce anxiety and foster emotional regulation. These formats prioritize feedback over judgment and are grounded in constructivist and humanistic pedagogies that emphasize continuous learning. Contemporary education must broaden its focus on the emotional dimension of learning, recognizing that school life extends far beyond anxiety alone. While this systematic review highlights the predominance of anxiety, it also reveals that academic and evaluative experiences generate a wide emotional spectrum, including enjoyment, hope, pride, shame, anger, hopelessness, and relief. Each of these emotions directly shapes how students learn, sustain motivation, and engage in relationships (Gao 2024; Gutiérrez-Vergara et al. 2020; Ávila-Toscano et al. 2021). As King and Frondozo (2022) observe, this dynamic reflects a diverse, changing, and necessary emotional landscape that molds students’ approaches to academic and personal challenges. Understanding this diversity requires integrative perspectives that consider biological, cognitive, social, and cultural dimensions, acknowledging both bodily reactions and the meanings students assign to their experiences (Weiner 2007). Interventions such as cognitive–behavioral strategies, emotion-regulation training, and formative feedback practices have shown positive outcomes in reducing anxiety and promoting emotions like pride and self-confidence (Furlan and Martínez-Santos 2023; Harley et al. 2021). These strategies highlight the potential for transforming assessment into a pedagogical opportunity for socioemotional growth.

In addition, the reviewed literature highlights several interventions that effectively transform students’ emotional experiences with assessment. Cognitive–behavioral interventions, for example, include guided relaxation, rational-emotive reflection, and self-instructional training. These strategies help reframe maladaptive beliefs and reduce anxiety, particularly in high-stakes or oral evaluation contexts (Furlan and Martínez-Santos 2023; Harley et al. 2021). Moreover, we identified three relevant categories of pedagogical strategies: (1) metacognitive and emotional self-regulation techniques, (2) socioemotional support systems such as mentoring programs and classroom dialog on emotional literacy, and (3) institutional-level reforms, including teacher training on emotional pedagogy and policy revisions promoting emotionally sustainable evaluation practices. These approaches bridge the gap between emotional theory and educational practice. Addressing this emotional spectrum not only helps prevent learning difficulties but, more importantly, creates opportunities to enhance the conditions that foster learning. In doing so, education can become more empathetic, supportive, and human-centered (Riaño-Rodríguez 2024; Gutiérrez-Vergara et al. 2020). A more human-centered education demands the dismantling of cultural and institutional barriers that prioritize standardized testing as the sole marker of academic quality. Without systemic shifts toward emotionally inclusive pedagogies, reforms risk remaining rhetorical rather than truly transformative.

5. Conclusions

This systematic review demonstrates that emotions constitute an essential, rather than ancillary, component of academic assessment processes. In the reviewed studies, the conceptualization of emotions and assessment (Q3) converges on defining emotions as complex, multidimensional, and contextual processes that are deeply interrelated with self-regulation, academic performance, and students’ identity formation. Assessment, in turn, emerges not only as a technical procedure, but also as an emotionally charged event, demanding a rethinking of its purposes and formats from a more humanized and pedagogical perspective.

The findings confirm anxiety as the predominant emotion in evaluative contexts, especially in standardized, oral, and memory-based assessments. By contrast, contextualized and collaborative formats foster positive emotions such as enjoyment, pride, and hope, which enhance learning (Q5). Most studies rely on quantitative and correlational methodologies, which, although useful for identifying patterns, limit a deeper understanding of emotional experiences in their subjective and contextual dimensions (Q4). This underscores the need to expand research through qualitative, mixed-methods, and longitudinal designs capable of capturing the complex dynamics of emotions in education. At the same time, the redesign of evaluative practices becomes imperative, with the integration of the emotional dimension as a formative axis to improve student well-being and educational quality. This study thus provides an updated framework on emotions in assessment, serving as a reference for researchers, teachers, and policymakers interested in developing more human, inclusive, and effective evaluation systems.

Our review therefore proposes an updated framework for emotions in assessment that integrates: CVT and its 2024 extensions (Pekrun 2006, 2024; Pekrun et al. 2011); self-regulation and coping as proximal mechanisms under evaluative stress (De la Fuente et al. 2020); discrete classroom emotions mapped to assessment formats (Goetz et al. 2010); test-anxiety mechanisms and interventions (Furlan and Martínez-Santos 2023); and measurement alignment drawing on foreign-language anxiety and ability-EI traditions. This synthesis offers a coherent map of constructs, instruments, and contexts to design more human, inclusive, and effective assessment systems.

Regarding influential authors (Q1), Pekrun (2024) and his control-value theory stand out, alongside contributions from Zeidner, Schwabe, and Furlan, whose work provides a robust framework for understanding emotions in assessment. This diversity of theoretical references reflects the interdisciplinary nature of the field, while also revealing the concentration of production within specific academic clusters. In relation to geographical distribution (Q2), the literature shows a strong concentration of studies in Latin America, particularly in Colombia, Ecuador, and Argentina, reflecting the regional concern with the emotional implications of assessment. However, the scarcity of studies in other cultural and educational contexts restricts the global generalizability of the findings.

As for key results and convergent conclusions (Q6), the studies consistently highlight that emotions exert a decisive influence on motivation, performance, and student well-being. Evidence also shows that targeted pedagogical interventions can transform negative emotions into positive ones, thereby improving both learning and emotional experience. Nonetheless, significant gaps remain regarding how to systematically integrate the emotional component into assessment practices and educational policy.

Ultimately, one pressing question remains open to educators, researchers, and policymakers alike: How can assessment systems be reconfigured so that they cease to be sources of anxiety and exclusion, and instead become spaces of recognition, equity, and the promotion of human potential?

Future Directions

Our review contributes an updated framework on emotions and assessment by integrating Control–Value Theory (including its 2024 extension) and the influential authors identified (Q1), clarifying conceptualization and the mechanisms of self-regulation and coping (Q3), mapping discrete emotions to assessment formats (Q5), and strengthening measurement via established traditions, while detailing the geographic distribution (Q2) and the dominant methodological profile (Q4). From this synthesis we identify gaps: the predominance of cross-sectional designs that limit causal inference (Q4), incomplete comparisons between formats and their emotional profiles (Q5), reliance on self-reporting with limited evidence of measurement invariance that reduces comparability (Q4), and regional concentration that restricts external validity (Q2). Based on these findings, we identify three main directions for future research and educational practice logically derived from the review: (1) designing emotionally sustainable assessment environments, grounded in evidence that standardized and high-stakes formats tend to elicit anxiety, while contextualized and collaborative formats foster enjoyment and pride (Zambrano-Vélez et al. 2023; Gao 2024); (2) integrating emotional literacy into teacher education, as studies show that teachers’ awareness and regulation of emotions directly influence students’ evaluative experiences and coping mechanisms (Gutiérrez-Vergara et al. 2020; Harley et al. 2021); and (3) reforming institutional assessment policies, given the persistent gap between research on emotional processes and the regulatory frameworks guiding evaluation practices (Riaño-Rodríguez 2024; Furlan and Martínez-Santos 2023). These priorities address the structural, pedagogical, and policy dimensions identified as key in our synthesis.

Accordingly, we propose longitudinal studies and, where feasible, randomized controlled trials that follow emotions before, during, and after assessment (Q4); experimental comparisons of standardized versus contextualized formats to estimate differential effects on anxiety, enjoyment, pride, and hope while modeling mediations by self-regulation and coping (Q5–Q3); multimethod measurement with consistency and invariance testing (Q4); and multisite, cross-cultural studies to expand external validity (Q2). This agenda operationalizes the updated framework and bridges evidence and school practice through brief, scalable interventions, yielding more robust and transferable conclusions (Q6).