Emotion understanding among institutionalized preschool children in India: A visual‐based approach

Abstract

The present study aimed to develop two visual tasks to assess the emotion understanding of institutionalized preschool children in India. To enhance the psychometric qualities of both tasks, content validity and inter‐rater reliability assessments, translation‐back translation and a rigorous peer review process were undertaken. In addition, the tool's components were mapped with institutionalized children's everyday experiences to help them better relate to the task. The tool development phase was followed by the assessment phase. Eighteen participants (nine males, nine females) aged 3–6 years were purposefully selected from three childcare institutions. The emotion identification task required children to identify the correct emotions from the cartoon characters' facial expressions and gestures, whereas, in the emotion situation task, children were asked to infer others' emotions from a vignette depicted by a picture card. Children's responses were videotaped, analysed and coded for both tasks. The study's key findings revealed that institutionalized children responded more accurately to emotion situation tasks yet struggled to identify sadness, anger and fear emotions compared to happiness. Furthermore, qualitative insights provided a comprehensive understanding and interpretation of the findings.

Key points.

What is Already Known on This Subject

• Children in institutional care are highly vulnerable, with numerous studies documenting the negative effects of institutionalization on overall development.

• Despite efforts to provide stimulation, nutrition, and healthcare, these children often face significant challenges in socio‐emotional development.

• Abused, post‐institutionalized children often struggle with emotion identification and comprehension tasks, typically demonstrating delays rather than deficits in understanding emotions.

What the Present Study Adds

• Highlights the necessity for contextually relevant assessment measures tailored for institutionalized children, particularly in India. Existing measures lack validation in this cultural context.

• Aims to create two visual tasks (emotion identification and situation tasks) using cartoon characters and picture cards to assess the emotion‐understanding abilities of preschool children in institutional settings through a child‐friendly approach.

• Emphasizes in aligning task scenarios with the everyday experiences of institutionalized children to enhance relatability and engagement.

• Involves rigorous peer review, content validity, inter‐rater reliability assessment, and translation‐back translation to improve the instrument's psychometric properties.

• Provides valuable insights into the emotion understanding of institutionalized preschool children in India, suggesting the potential for tailored intervention strategies.

INTRODUCTION

Children require a compassionate, loving and supportive home environment in order to flourish and mature harmoniously (United Nations Convention on the Rights of the Child, 1989). Unfortunately, millions of children worldwide have been separated from their families for various reasons, including but not limited to seeking protection and for social and economic reasons (E.C. Daphne Programme, 2005). These children are typically raised in alternative care settings for varied periods until they are adopted or reunited with their family or, in some situations, until adulthood. In many low resources and developing countries in Asia and Africa, poverty and parental instability, including parental illness, are considered significant factors for the institutional placement of children (Bunkers et al., 2014). In India, ‘approximately 1.8 lakh children are in institutional care because their parents are unable to provide for their basic needs’ (Jena Committee Report, 2017).

Impact of institutionalization on children's development

Most research on institutionalized children is conducted in Western countries, while there is a paucity of data in the Indian context. Extant studies on institutionalized children acknowledged such children to be incredibly vulnerable, and ‘institutionalization has been associated with significant delays in all aspects of children's development, including physical, behavioral, cognitive, social, and emotional’ (Carter & Everychild, 2005; Johnson et al., 2006). Even though childcare institutions differ in their approach and practices regarding their functioning across and within countries, the following can be summed up as a generic representation of ‘what is typical’ across the institutions (Rosas & McCall, 2011). A few examples are large group size, low caregiver–child ratio, inconsistent and unstable caregiving patterns by inexperienced caregivers with little or no training and lack of meaningful caregiver–child interaction with little warmth, sensitivity and responsiveness to the children's emotional needs. It has been demonstrated that despite institutions' best efforts to provide them with various forms of stimulation, nutrition and health care, these children continue to struggle with socioemotional development (Hodges & Tizard, 1989).

Children raised in institutional care often lack the caring and stimulating environment required for optimal psychological development. They receive minimal communication and attention from caregivers, and their individual needs are rarely met (Johnson, 2000; Rutter & The English and Romanian Adoptees Study Team, 1998). Moreover, institutionalized children frequently exhibit substantial developmental delays and may follow unconventional developmental trajectories (Van IJzendoorn et al., 2011). Existing research suggests that the majority of children in institutional care have negative experiences that impact their development (Abdullah et al., 2018). In this regard, the work of Goldfarb (1943) and Bowlby (1951) was particularly influential, highlighting the emotional, behavioural and cognitive impairments observed among these children. Furthermore, these children were often reported to have intellectual delays and specific difficulties with language development. They also struggled with concentration, forming emotional connections and frequently exhibited attention‐seeking behaviours.

A recent meta‐analysis encompassing more than 300 studies and over 100,000 institutionalized children from 65 countries has identified significant delays in their physical growth, brain development, cognition, attention and an increased prevalence of atypical or disorganized attachments. The analysis also revealed a two‐fold rise in insecure attachment and a four‐fold increase in disordered or unclassifiable attachments (Van IJzendoorn et al., 2020). Numerous researchers have claimed that these children have limited scope to form stable and ongoing attachment relationships due to inadequate and poor‐quality interaction with their caregivers (Vorria et al., 2003; Zeanah et al., 2005).

In addition, institutionalization can also have a detrimental impact on a child's physical development and health due to inadequate nutrition and limited access to medical care. Research has shown that children in overcrowded and understaffed institutional settings in India often exhibit stunted physical growth, with their weight, height and head circumference falling nearly two standard deviations below the norm and their average IQ being somewhat lower than 70 (Juffer et al., 2017). Moreover, these children frequently exhibit delayed cognitive development compared to their non‐institutionalized counterparts. Hodges and Tizard (1989) found that such children exhibit more behavioural issues than those raised in family environments. A meta‐analysis by van IJzendoorn et al. (2008) indicated that institutionalized children exhibit poorer social skills and higher rates of social maladjustment. These children were also found to experience greater emotional problems (Erol et al., 2010; Ford et al., 2007), exhibiting emotional withdrawal (Zeanah et al., 2005) and emotional loneliness (Han & Choi, 2006; Ptacek et al., 2011).

Emotion understanding: the developmental trajectory

Emotion understanding encompasses a multifaceted process that involves a diverse range of skills, including emotion identification, emotion knowledge (Castro et al., 2016), the comprehension of emotion categories and concepts and the accurate interpretation of others' emotions within a cultural and social context (Denham et al., 2016). The essential skills for developing emotion understanding begin early in life, starting in infancy (Denham, 1986). It extends throughout childhood as children progressively grasp the nuances inherent in subtle and ambiguous emotional occurrences. Nonetheless, children's initial grasp of basic categorical emotions serves as a fundamental framework for a more elaborate understanding. During preschool, typically developing children learn to identify different emotional expressions and comprehend the underlying reasons behind emotional responses (Southam‐Gerow & Kendall, 2002). By age three, most children can predict another individual's emotional state by associating a puppet's facial expression with a corresponding vignette (Denham, 1986).

Impact of adversity on emotion understanding

Although scarce, research on vulnerable children has well documented the adverse and unusual experiences during the formative years to be associated with abnormal emotion processing. Similarly, when compared to non‐abused children, abused children demonstrated less developed emotion identification skills, such as poor recognition of different facial expressions, particularly happiness and a lack of understanding of the causes of emotions (Camras et al., 1983; Pollak et al., 2000). In addition, abused children displayed bias in their responses when asked to match facial expressions to situations that elicited anger (Pollak et al., 2000). Compared to their counterparts, they needed less perceptual information and displayed distinct category boundaries in identifying anger rather than happy, sad and fearful expressions (Pollak & Kistler, 2002).

Within a structured lifestyle and early neglect, institutionalized children encountered atypical exposure to faces and facial emotions. They came across numerous unfamiliar faces, fewer instances of their primary caregiver's face and a limited range of positive and negative facial expressions during their stay. As a result, their exposure to facial expressions became restricted, depriving them of opportunities to learn how their behaviour can elicit various facial responses (Moulson et al., 2015). In a comparative study, Moulson et al. (2015) discovered that institutionalized children had a higher threshold for identifying happy emotions than sad, angry, or fearful expressions compared to foster care and community‐based children.

Similarly, various studies have been undertaken in the United States and Europe to investigate the effect of adverse institutional experiences on the emotion understanding of post‐institutionalized children. When asked to match the facial expressions for happy, sad and fearful vignettes, Fries and Pollak (2004) found that post‐institutionalized preschool children struggled with emotion identification and situation tasks. Despite performing poorly on these tasks, they demonstrated developmental trajectories similar to that of community children in the emotion identification task. Moreover, a similar kind of study was conducted by Tarullo et al. (2007) to assess 6‐ to 7‐year‐old post‐institutionalized children's emotion understanding abilities by using three computerized tasks (emotion identification, emotion situation and emotion antecedent task) across three groups (children adopted into institutional care, foster care and non‐institutionalized children). Surprisingly, the outcomes of this study revealed that post‐institutionalized children performed no differently on the three emotion tasks than the other two groups. According to the researchers, one possible reason for these contradicting findings is that, since the activities were primarily developed for younger children, the older children surpassed all of them, resulting in fewer missing and more accurate responses. However, both studies found that post‐institutionalized children may experience a delay rather than a deficit in comprehending emotions.

Emotion socialization among young children in India

The process of emotion socialization seeks to develop children's emotional competence. Cultural factors often influence this process, shaping caregivers' expectations regarding the appropriateness of children's emotional expression and experiences. In White American families' settings, where most emotion socialization research is undertaken, parental goals tend to promote personal choice, self‐expression and independence. In contrast, the Indian socialization process is driven by a deep desire for group and familial integration (Panda & Gupta, 2004). Parental socialization goals in Indian families emphasize training children to maintain mutually beneficial connections while respecting their autonomy (Tuli & Chaudhary, 2010).

Furthermore, India's cultural ideals emphasize the experience and expression of specific emotions. For example, happiness is frequently regarded as an immature emotion, only valid for young infants who are still unaware of life's facts and responsibilities (Menon, 2000). This view is replicated in urban Indian contexts, where mothers deliberately regulate their children's happiness and empathy (Kathuria, 2019). Further, the Indian cultural context emphasizes the complexities of group association and individual positioning, which are heavily influenced by specific interactions and contexts (Chaudhary & Sriram, 2020), implying the significance of hierarchical relationships as a key determinant of the propriety of emotion expressions. For example, fear for the father was always promoted in India, but anger was discouraged (Pai, 1998). Furthermore, children in India were less likely to display anger in the presence of their mothers than peers and less inclined to express sadness and pain in the presence of their fathers than mothers (Raval et al., 2007). In addition, studies have shown that girls are more inclined than boys to mask their emotions, implying that anger is more socially acceptable in boys (Raval et al., 2007).

The majority of empirical studies on emotion socialization in India have included both mothers and children. According to the findings, mothers with relationship goals prioritize providing their children with precise instructions (training) for negative disengaging emotions, while other caregivers frequently advocate for comforting reactions to negative, socially engaging emotions since they do not endanger group or interpersonal ties (Kathuria et al., 2022). Additionally, Raval and Martini (2009) found a positive correlation between the children's self‐reported control of anger, sadness and pain and the urban middle‐class Hindu mothers' reports of punitive and minimizing behaviours (which are typically associated with negative child outcomes in the West). In the Indian context, these results could be deemed advantageous. In contrast, no relationship was found between Indian children's self‐reported emotion regulation and mothers' accounts of soothing and problem‐solving activities, which are generally seen favourably in the West. Apart from the problem‐solving reactions to the emotions of the children (which are characteristic of middle‐class white mothers), Indian women often explain the situation to the child to ‘make the child understand’. This technique is consistent with the broader relational socialization goals, which enable children to adapt their own goals and needs to fit their social environment and accept the circumstances, thus fostering their development into socially integrated adults.

The current body of research primarily focuses on non‐institutionalized children, leaving a significant gap in understanding the emotion socialization process among institutionalized children. Furthermore, the existing studies have been predominantly conducted in Western India, with limited data available from other regions of the country. Given the cultural and regional diversity within India, it is crucial to examine the role of caregivers in the emotion socialization of institutionalized children, as they serve as the primary conduits for transmitting cultural values through the socialization process. Ultimately, this will contribute to a deeper, culturally informed understanding of the developmental processes through which institutionalized children's emotions are socialized and the pivotal role the caregivers play in this process.

The present study

The institutionalization of children often has detrimental effects on their overall well‐being, prompting child rights advocates, including international recommendations on alternative care (United Nations, Guidelines for the Alternative Care of Children, 2009), to emphasize the importance of closing all childcare institutions and establishing high‐quality foster care programmes. While some nations are transitioning from institutional to foster care, a substantial number of children, particularly in low‐income countries (McCall, 2011, 2013), like India, continue to reside in institutional settings due to the growing demand for such care and the limited progress in developing robust foster care systems. Additionally, cultural and religious beliefs in India often refrain people from adopting someone else's child, further limiting alternatives to institutional care. Compounded by the prevailing preference for adopting infants and younger children, older and disabled children make up a significant portion of India's institutional care facilities (Mishra & Sondhi, 2018). Therefore, the present study focuses on examining the emotion understanding of institutionalized preschool children. Gaining insight into their emotion understanding abilities will be crucial in understanding their emotional development.

A thorough review of the literature showed that only two studies have been conducted in the United States and Europe so far to investigate the emotion understanding ability of post‐institutionalized children, revealing a glaring dearth of research in this area. Nevertheless, scant attention has been devoted to exploring the emotion understanding ability of institutionalized preschool children within the Indian context. Moreover, the existing measures for assessing institutionalized children's emotion understanding through various tasks have not been validated in the Indian context. Also, some elements of the previously used tasks may not be suitable in the institutional context due to cultural and contextual differences. Therefore, to address these gaps, the current study aims to develop two visual tasks (emotion identification and emotion situation tasks) to assess the emotion understanding of institutionalized preschool children aged 3–6 years in India. The age range was carefully chosen because these are the critical and formative years of development when children start learning many emotion‐understanding skills that aid them in understanding, expressing and socializing their emotions as they grow.

MATERIALS AND METHODS

Research design

The current study employed an exploratory design. Data were collected and analysed primarily using qualitative tools and techniques followed by descriptive analysis.

Participants

Participants for this investigation were selected through purposive sampling. Explicit inclusion and exclusion criteria were established to guide the recruitment process. A total of 18 institutionalized children (nine males and nine females) aged between 3 and 6 years who satisfied the inclusion criteria were selected from three distinguished Childcare Institutions (CCIs) within Odisha (Eastern part of India) operated by non‐governmental organizations. All the three CCIs included in the study were situated in urban areas.

Out of the 18 participants, two had experienced abuse, seven were placed in the institution due to parental neglect, two were runaway children and seven were institutionalized because of their mother's mental illness. The present study did not include children in conflict with the law, gifted children, differently abled children, adopted children or foster care children. Moreover, children with speech delays, language difficulties or hearing issues were excluded from the study. This was to ensure that the participants' limited verbal abilities did not interfere with their performance across both tasks. Children below the age group of 3 years and above 6 years were also excluded from the study. In the emotion situation task, only one child refused to comply, while all 18 children participated in the emotion identification task. All the children included in the study came from a low socioeconomic background.

Ethical considerations

The study underwent a rigorous ethical review process, obtaining multiple layers of approval. First, the study was thoroughly reviewed and approved by the appropriate Institute Ethics Committee (NISER/IEC/2022‐06). Second, official approval was sought from the Chairpersons of the respective Child Welfare Committees (CWCs) to access various CCIs under their jurisdictions. Third, the gatekeepers, namely, the superintendents of the respective CCIs, willingly extended their consent for the study. Finally, because the study participants were minor, vulnerable children, informed agreement was obtained from their caregivers. Along with informed consent, the institutionalized children provided assent to participate in the study. Throughout the research process, utmost confidentiality and privacy were maintained, safeguarding the participants' identities and the CCIs they were associated with. Each child was assigned a code to ensure anonymity.

Tool development process

The tool development process encompassed various phases, including item generation, creating cartoon characters and picture cards for both tasks, rigorous peer review, content validity assessment, translation‐back translation and pilot testing.

Item generation

The tool consisted of two tasks: an emotion identification task and an emotion situation task. The items for both tasks were generated following a thorough assessment of relevant literature and referring to already established measures. Many of these measures, including the Florida Affect Battery (Bowers et al., 1991), Facial Affect Comprehension Evaluation (Mrakotsky, 2001), Facial Expression Recognition measure (Pons et al., 2002) served as references to design the items for the emotion identification task. All these measures used only facial expressions to understand the emotion recognition ability of typical preschool‐aged children. For depicting the facial expressions, either hand‐drawn photographs of both child and adult characters or animated digitized photographs were used. On the other hand, for the emotion situation task, different vignettes in the form of stories as well as hand puppets portraying different emotion situations, were utilized to assess the emotion understanding ability of non‐institutionalized preschool children (Denham, 1986).

Additionally, multiple visits to various CCIs and discussions with staff members were undertaken to gain a comprehensive understanding of the environment in which institutionalized children resided and to understand their everyday experiences. The components of the tool were mapped with the institutionalized children's everyday experiences, aiming to establish better relatability to the scenarios presented. A total of 33 items were generated for both tasks. Rather than adapting any items from the previously established measures, the researchers tailored the items to suit the specific cultural and institutional context. A total of 20 items for the emotion identification task and 13 items for the emotion situation task were generated.

Creating cartoon characters and picture cards for both tasks

The present study devised visual tasks to elicit responses from the participants. This was a qualitative way of acquiring data using visual materials (Creswell & Clark, 2017, p. 111). A professional artist was enlisted to create cartoon characters for the emotion identification task and picture cards for the emotion situation task. After designing the cartoon characters and picture cards, it underwent another rigorous peer review process, during which numerous modifications were made to improve the items' accuracy and clarity. This qualitative technique was applied to enhance the psychometric properties of the instrument.

Three potential reviewers, consisting of a preschool educationist, a developmental psychologist and a psychologist from a childcare institution other than the experts who conducted the content validity assessment, were invited to evaluate and ascertain the suitability of the cartoon character's facial expressions and gestures for the emotion identification task, as well as the vignettes and picture cards employed in the emotion situation task. The reviewers were requested to provide detailed feedback on whether the items in both tasks aligned with the everyday experiences of institutionalized children and whether they were easily comprehensible, particularly in terms of the wording and story content. Based on their recommendations, several modifications were made, including adjustments to cartoon character's facial expressions and gestures to enhance accuracy. In addition, modest changes were made to the picture cards by minimalizing the background objects and adding necessary details to improve understanding by institutionalized children. Additionally, the story content was refined using simple, age‐appropriate and familiar language for better relatability and connection.

Content validity assessment

To assess the content validity of the developed tool, a panel of nine experts with expertise in relevant domains was contacted, including clinical child psychologists, developmental psychologists, experts having knowledge and expertise in qualitative research methodologies, scale development, projective techniques and people working in various CCIs. These experts possessed more than 10 years of extensive experience in their respective fields. In contrast, people working in various CCIs had more than 5 years of experience working with institutionalized children.

For both tasks, content validity was assessed using the Lynn (1986) technique. The content validity index (CVI) is a commonly used method of evaluating a tool's content validity. The CVI for both tasks was calculated at the item level (I‐CVI) and for the entire instrument (instrument level‐CVI). The I‐CVI was calculated using expert judgements of each item's content representativeness and relevance on a 4‐point Likert scale ranging from 1 (not representative or relevant) to 4 (extremely representative or relevant). The I‐CVI was computed by dividing the number of experts rating the item as 3 or 4 by the total number of experts, yielding the percentage of agreement on the item's content validity. The Ave‐CVI (average proportion of items rated as 3 or 4 across the various experts) was the summation of the I‐CVIs divided by the number of items. In addition to CVI, the modified kappa (k*) was also calculated to adjust each I‐CVI for chance agreement (Polit et al., 2007).

For the content validity assessment of the emotion identification task, 20 items were included, of which 17 were retained and three were discarded due to a CVI of less than .78 (Lynn, 1986). Similarly, based on the CVI scores, 11 out of 13 items were maintained for the emotion situation task. The computed CVI for both tasks was .81, and the modified kappa for all items in both tasks was (K ≥ 0.74), which is excellent.

In addition, modifications were made in response to the expert's qualitative inputs and suggestions. For instance, in the emotion identification task, the facial expressions and gestures of the cartoon characters were modified several times to improve clarity and accuracy. Similarly, the items were slightly revised for the emotion situation tasks, such as rewording items to fit the institutional context. In addition, in response to the expert's recommendations, the picture cards underwent minor revisions for better and easier understanding by the institutionalized children.

Translation and back‐translation

Due to the language barrier between the tool, initially developed in English, and the target population's comprehension of the regional language, translation and back‐translation were performed. Two translators translated both tasks into the regional language (Odia), and afterward, two other translators, fluent in English, back‐translated it. This process ensured quality assurance, and the resulting reports from the translators were integrated into one report during the synthesis phase. Finally, an expert panel consisting of a methodologist, a language specialist and a translator reviewed all the reports, providing comments and recommendations for preparing a final document.

Tools used

Emotion identification task (description and administration procedure)

The emotion identification task aimed to understand the institutionalized children's ability to recognize emotions through cartoon characters' facial expressions and gestures. A total of 17 cartoon characters (four showing happiness, four showing sadness, four showing anger and five showing fear) were introduced in a counterbalanced manner to prevent order effects. For each item, four response options were given, including one correct and three incorrect choices, and the children were asked to determine the emotion experienced by the character and whether the character appeared happy, sad, afraid or angry.

The exact instructions for all children in the emotion identification task are as follows: ‘Let's play a guessing game. It would be so much fun and exciting. Do you enjoy watching cartoon characters? Well, I will show you different cartoon characters, and we will play with them. Now, look carefully at their face (whether they look happy, sad, scared, or angry) and gestures (their hand and leg positions) and tell me how the cartoon character feels in each picture. For each cartoon character, I will provide you with four options (happy, sad, angry, or scared). Then, you need to choose the option that best matches the cartoon character's facial expressions and gestures. There are no right or wrong answers. You are free to express how the cartoon character is feeling. Let me first show you how to play’.

Emotion situation task (description and administration procedure)

An emotion situation task was also administered to evaluate the institutionalized children's ability to understand others' emotions through hypothetical emotion‐eliciting situations known as vignettes. Eleven short vignettes depicting different emotions, including happiness (three vignettes), sadness (two vignettes), fear (three vignettes) and anger (three vignettes), were presented. Each vignette involved a picture card illustrating a hypothetical emotional scenario. These picture cards proved to be quite useful, as they engaged the children's visual and auditory senses. Most of the vignettes in the emotion situation task featured only two characters (the caregiver and the child), while few contained three characters (caregiver, child and friend). Furthermore, to avoid difficulties and recognition caused by age and gender, the characters in the picture cards were similar in age to the participants. Also, the emotion situation task included an equal representation of male and female characters under each emotion but not necessarily matched the gender of the participants. Moreover, to portray the story in a simplistic manner, the minimum background details were added to avoid potential distractions. The picture cards featured a simple drawing with only the characters and the necessary objects, enabling the researcher to point to the characters and provide detailed descriptions of their activities before narrating each story.

To add a level of challenge and prevent any ceiling effects, only the neutral facial expressions of the characters were visible to the children. Moreover, the emotion labels were also not provided in the story. This made the task slightly challenging for the institutionalized preschool children, prompting them to engage in thoughtful consideration before responding. As a result, the children had to infer other's emotions from the stories depicted by a picture card.

Furthermore, the researcher took measures to verify the children's complete understanding of the scenarios. The children were given sufficient time to process the information and were instructed to paraphrase it in their own words. This approach facilitated their capacity to raise questions and allowed the researcher to effectively address any doubts or uncertainties they had. Then, the children were asked to determine the emotions felt by the depicted character (either a child, caregiver or friend) in each picture. Moreover, they were given four alternatives (happy, sad, angry and scared) for each story to help them better understand the scenarios and respond effectively. They were also requested to label the emotion while responding to the question.

The instructions for the emotion situation task are as follows: ‘I have another fun game for you. Do you enjoy listening to short stories? In this game, I will show you different picture cards and talk about the characters in each story. I will discuss each character's activities (what they are doing) in the story. If you have difficulty understanding something, please let me know. I will do my best to help you understand. Please listen carefully and understand the stories in each picture card. Then, please tell me how the character feels in each picture in the story. I will give you four options (happy, sad, angry, or scared) from which you can choose. Let me first show you how to play’.

Standardization of procedures for emotion identification and situation task

Significant efforts were invested to ensure the children's understanding by providing clear, age‐appropriate instructions using simple, familiar words. A standardized script was prepared, comprising all relevant details such as instructions, examples and response format, which the researcher consistently adhered to across all participants. The instructions were presented one at a time, allowing the children to fully process the information and ensuring that they clearly understood the instructions. The researcher attempted to maintain a consistent pace, tone of voice, facial expressions, body language and pauses throughout the data collection process. Additionally, the researcher remained neutral and objective, even when the children gave incorrect responses. Significant emphasis was placed on presenting the items for both tasks in an identical manner for all the participants. Furthermore, the items and response choices for each child were presented in a counterbalanced order to prevent any order effects for both tasks.

Data collection

Due to difficulties in accessing institutionalized children for the pilot study, the emotion identification and situation tasks were instead carried out with 15 non‐institutionalized children aged 3–6 years, with some modifications. Informed consent was obtained from their parents and assent was taken from the children. The characters were introduced as parental figures and siblings instead of caregivers and friends, and the administration procedures were the same. The pilot study was purposefully undertaken before the field study to identify any potential areas of concern early in the research process, including the length of data collection, the wording used, the feasibility of the tool and the challenges faced by participants while responding.

Following the pilot study, multiple visits were made to various CCIs at scheduled times based on the availability of the children and prior consultations with the management personnel. Before the study commenced, substantial effort was put into establishing a positive rapport with the children. This involved spending approximately 10 days at each institution, actively engaging with the children in enjoyable activities such as playing, storytelling, showing YouTube videos and offering small gifts. The intention was to create a comfortable environment where the children could feel at ease. Before administration, it was ensured that the children were neither bored, fatigued, nor in an unpleasant mental state that could impair their task performance. Each task was administered individually to each child in a separate room by the first author. Even though inclusion and exclusion criteria were predetermined, certain aspects of both tasks were administered to the excluded children to prevent them from feeling excluded. Both tasks were administered in the regional language (Odia), which was the mother tongue of the participants and the first author. This linguistic choice facilitated a smooth and seamless interaction with the children. The emotion situation task followed the emotion identification task. At the beginning of each task, the children were provided with examples to familiarize themselves with the procedures. Throughout the research process, the children's well‐being was the top priority. As part of this effort, the children were explicitly informed that no definitive or incorrect responses exist. It was emphasized that they should be able to convey their thoughts freely to the researcher without being perceived as incorrect or subject to judgement.

The children were evaluated on two separate occasions, 1 week apart, with each session comprising one task only: emotion identification or emotion situation assessment. This was purposefully done, considering their short attention spans. The interval between the two tasks was standardized and maintained consistently across all participants throughout the investigation, thereby eliminating potential variations between participants and sessions. Each task took approximately 10–15 min to administer. The children's verbal and behavioural responses were videotaped for subsequent data analysis.

Data analysis

Qualitative data analysis

Following data collection, all video recordings underwent a comprehensive review by the first author. The children's responses were meticulously analysed before assigning codes for the emotion identification and situation tasks. The coding criteria were established earlier based on the objective responses in both tasks. In contrast to open‐ended questions, the children were posed with specific close‐ended questions, which could be answered by selecting a specific response from the alternatives provided. For the emotion identification task, after presenting each cartoon character, the children were asked, ‘How does the character feel in each picture?’. They were offered four choices (happy, sad, angry or scared) to select from. Likewise, in the emotion situation task, after displaying each picture card and explaining the activities of the characters in detail, the children were asked the following question, ‘How does the character feel in this story?’. The children were given the same options (happy, sad, angry or scared) to choose from. This approach enabled a standardized analysis of the data across participants throughout the research process.

Furthermore, the coding patterns were reviewed by experts during the content validity assessment phase, and a consensus was reached among all the experts on the predefined codes. As per the assigned codes, the children's responses to each item were categorized as correct, incorrect or confused for both tasks. Even with prompting, the children's responses were classified as correct if they identified and understood the emotions appropriately. However, if the children provided incorrect answers despite repeated prompting and alternatives, their responses were coded as incorrect. Finally, if the children answered with more than one option, repeated the last option without understanding and appeared confused and unsure, their responses were labelled as confused.

In addition to coding, detailed notes were taken on the qualitative inputs from the in‐depth analysis of the video recordings. These notes included the frequency of alternatives provided, the number of prompts used to elicit responses, the children's level of confidence in identifying and understanding various emotions and their response times. The findings from the qualitative analysis are discussed in the Results section.

Descriptive analysis

The analysis predominantly relied on qualitative approaches; however, descriptive statistics were used only to aid in summarizing and explaining the performance of institutionalized children across two different tasks. After the initial coding, the total number of correct, incorrect and confused responses was calculated emotion‐wise, and descriptive statistics were generated in Excel to summarize the performance in both tasks. The data were visually represented using clustered bar diagrams.

Data trustworthiness

Because the first author collected and analysed the data, two independent coders thoroughly reviewed, independently analysed and coded each child's performance in both tasks to ensure data trustworthiness and minimize potential researcher bias. In cases where disagreements arose between the first author and the coders, the second author intervened to cross‐check the data and made the final decision. Moreover, an inter‐rater reliability analysis was performed to quantify the agreement between two coders by using Cohen's Kappa. For the analysis, the DATAtab online statistics calculator was used (DATAtab Team, 2024). For the emotion identification task, the findings showed an almost perfect agreement between two coders (κ = 0.84). On the other hand, for the emotion situation task, Cohen's Kappa revealed a substantial agreement between two coders (κ = 0.8). In both tasks, the p‐value was less than the conventional threshold of .05, indicating that there is a low probability that the observed agreement is due to random chance.

RESULTS

Both the qualitative and quantitative findings from the emotion identification and situation tasks are shown below.

Qualitative findings

Emotion identification task

Out of 18 children included in the study, 14 of them were required to be given one correct and three incorrect choices for each item, while four children responded more confidently and quickly without the options. Regarding children's response time, except for the abovementioned four children, the rest of the 14 children took comparatively longer to respond and required significant prompting for a reasonable amount of time. Furthermore, the same 14 children required indications, such as repeated signals to gaze at the cartoon character's facial expressions and gestures, to respond. It was found that three of these children merely repeated the researcher's last option without correctly identifying different emotions.

One of the intriguing findings was revealed regarding labelling emotions, especially happiness and sadness. It was observed that some of the children struggled to label the emotions as happy or sad. However, they responded quickly when asked if the cartoon characters in the task were smiling or crying. Furthermore, they utilized various local languages to convey these emotional states, such as the cartoon characters feeling ‘okay’ for happiness and ‘not okay’ for sadness. Despite being presented with four distinct options (happy, sad, anger and fear) for each item, the children were unable to accurately select the appropriate label, opting instead to convey happiness and sadness through the general descriptors of ‘feeling good or bad’.

Moreover, the researcher sought to provide more comprehensive prompting when children struggled to recognize various emotions, even after providing repeated prompting and alternatives. For instance, they were given contextual cues to facilitate understanding, such as ‘Imagine your caregiver gave you your favorite candy’. ‘Could you please express your feelings on your face?’ Then, the children were instructed to relate the same with the cartoon character's facial expressions and gestures in the emotion identification task. In certain cases, these contextual cues proved effective in eliciting responses from the children.

Emotion situation task

Compared to the emotion identification task, the emotion situation task was relatively more challenging as it required the children to infer the emotions of others from the vignette depicted by a picture card. In addition, the characters in the picture cards were emotion‐neutral for happy, sad, angry and fearful situations. The qualitative inputs from the video recordings highlighted that institutionalized children found the emotion situation task less complicated and responded more quickly and appropriately. They took relatively less time to respond than the emotion identification task. Even though the children were provided with the options for each item, they responded with little or no prompting. In situations involving anger, five out of 17 children responded ambiguously and incorrectly. A careful examination of the video analysis revealed that these five children mistook anger for sadness. They stated that they would be more sad than angry if their friend took their toy without asking or dismantled their house made of building blocks for no reason. A more detailed analysis of the children's responses to anger‐inducing scenarios revealed that their interpretations cannot be deemed entirely erroneous; a cultural perspective may provide valuable insights into this phenomenon. Because different cultures prioritize certain emotions over others (e.g. sadness vs. anger), the cultural context can influence caregivers to emphasize emotions that are more significant within their society. A detailed discussion of potential cultural reasons for this observation can be found in the discussion section. Furthermore, in situations involving sadness, almost all the children correctly understood the emotional situations, followed by happiness. However, these children gave more incorrect and confused responses in fearful than happy and sad situations.

Quantitative findings

Figures 1 and 2 present a descriptive analysis of emotion identification and emotion situation tasks for both institutionalized and non‐institutionalized children. Additionally, Figure 3 displays the average total scores, including error bars, attained by institutionalized boys and girls across two distinct age groups (3–5 and 5–6) in the emotion identification task.

FIGURE 1.

Descriptive statistics for emotion identification task among institutionalized and non‐institutionalized children. A clustered bar diagram illustrating the percentages of correct, incorrect and confused responses provided by both institutionalized and non‐institutionalized children in the emotion identification task under happiness, sadness, anger and fear category (Author's source, Excel data analysis).

FIGURE 2.

Descriptive statistics for emotion situation task among institutionalized and non‐institutionalized children. A clustered bar diagram showing the percentages of correct, incorrect and confused responses provided by both institutionalized and non‐institutionalized children in the emotion situation task under the happiness, sadness, anger and fear category (Author's source, Excel data analysis).

FIGURE 3.

The average total score of institutionalized children in emotion identification task by age and gender. A clustered bar chart illustrating the average total scores (along with the error bars) attained by both institutionalized boys and girls across two distinct age groups (3–5 and 5–6) in the emotion identification task (Author's source, Excel data analysis).

Lastly, Figure 4 represents the percentage of accurate responses of institutionalized children in both emotion identification and situation tasks.

FIGURE 4.

Performance of institutionalized children across emotion identification and emotion situation tasks. A cluster bar diagram displaying the percentage of accurate responses of institutionalized children under the happiness, sadness, anger and fear categories in both emotion identification and situation tasks (Author's source source, Excel data analysis).

DISCUSSION

The present study delved into the emotion understanding ability of institutionalized preschool children aged 3–6 years in two fundamental emotion tasks. The rationale for evaluating these tasks at this specific age is their foundational role in more intricate social interactions. In conjunction with descriptive analysis, the utilization of qualitative data collection methods and thorough video analysis contributed to a comprehensive understanding and interpretation of the findings.

The descriptive analysis of the emotion identification task unveiled that institutionalized children demonstrated a higher degree of accuracy in identifying happiness, followed by sadness, anger and fear (refer to Figure 1). A closer examination of the percentage of correct responses indicated that, although happiness garnered the highest level of accuracy, the percentage was slightly above average. Conversely, accurate responses for sadness, anger and fearful facial expressions were nearly above average and slightly below average. Thus, it can be inferred that without contextual cues, these children faced difficulties identifying sadness, anger and fear emotional expressions, as opposed to happiness. Furthermore, the institutionalized children exhibited greater ambiguity in identifying fearful emotional expressions compared to sadness, happiness and anger. The overall findings of the emotion identification task are consistent with the previous research findings with post‐institutionalized children, indicating that these children had significant difficulties in identifying various emotional expressions (Fries & Pollak, 2004).

In addition, the aforementioned findings also significantly contradict the outcomes reported by Moulson et al. (2015) regarding the identification of positive versus negative emotions (e.g. happiness vs. sadness, anger and fear), showing that institutionalized children exhibited higher thresholds for recognizing happy facial expressions compared to sad, fearful and angry faces. Furthermore, these children demonstrated lower sensitivity and required more perceptual information to differentiate happy faces from negative emotional expressions. In contrast, our study revealed that institutionalized preschool children identified happiness more accurately than other emotions. Although we do not possess independent data to substantiate this claim, informal discussions with caregivers of institutionalized children provided valuable context for interpreting these findings. Caregivers have expressed a belief that negative emotions, such as anger and fear, may hinder the children's developmental progress; thus, they often discourage the discussion and expression of such emotions. Consequently, it is plausible that caregivers in different CCIs intentionally emphasize the incorporation of positive emotions in daily interactions with the belief that fostering positivity benefits the children's well‐being. This explanation may help elucidate the observed trend of institutionalized children displaying greater accuracy in identifying happiness over other emotions.

Moreover, the findings of the same study by Moulson et al. (2015) also indicated a significant disparity in emotion recognition capabilities between institutionalized children and those raised in family settings, particularly regarding the identification of happiness. Notably, the study found no significant differences in the recognition thresholds for sad, fearful or angry facial expressions across the two groups. This led the authors to suggest that early social deprivation does not significantly impede perceptual sensitivity to negative emotions. In our investigation, we observed that non‐institutionalized children demonstrated greater accuracy in recognizing a diverse range of emotional expressions (both positive and negative) compared to their institutionalized peers. While the statistical analysis was constrained by the qualitative design of our study and the limited sample size, we postulate that the restricted exposure to a diverse range of facial emotional expressions within institutional settings, as opposed to the richer emotional context typically present in family environments, may elucidate the observed discrepancies in emotional recognition proficiency. This hypothesis is further substantiated by informal discussions with caregivers of the institutionalized children, whose observations are captured in the following verbatim excerpts.

Informal discussions with the caregivers and management staff of the CCIs were also carried out to understand the reasons behind certain children's incorrect and confused responses. When asked about the children's knowledge and understanding of basic emotions and whether they label them in their daily lives, one of the caregivers stated, ‘Due to their young age, children understand happiness and sadness through smiling and crying, rather than explicitly labeling them as such. They rarely use these labels and instead learn from others' facial expressions and outward display of emotions’. On the other hand, when questioned about why some children could not assign labels to emotions, their caregivers responded, ‘Since we have never used these terms in their presence, the children are unfamiliar with these labels. Additionally, some children are not enrolled in a school where they would have had the opportunity to learn about these emotions. Our institutions lack the necessary resources and infrastructure for providing emotional education’. Another caregiver stated, ‘Even though our institution has a school on its premises, these children have not been taught about various emotions and feelings’.

In addition to the quantitative analysis of the emotion identification task, qualitative insights derived from detailed observation of video recordings showed that children in the age range of 5–6 were more proficient in appropriately identifying emotions with minimal or no prompting. In contrast, younger children required more time and prompts to respond. This is further supported by the descriptive statistics (refer to Figure 3). During the emotion identification task, children below 5 years encountered challenges in assigning labels to the emotions portrayed by the cartoon characters, mainly happiness and sadness, and struggled to apply these emotional terms. However, they responded swiftly when asked whether the cartoon characters in the task were smiling or crying. Therefore, it could be assumed that this restricted emotional vocabulary hindered their ability to label the emotions depicted by the cartoon characters. Additionally, it was observed that the institutionalized children demonstrated a relatively greater familiarity with the labels of anger and fear compared to happiness and sadness. Moreover, for most children, repeated prompts and alternative responses were necessary to elicit answers, as they struggled to respond without prompting.

Numerous studies on non‐institutionalized children emphasized the importance of a family environment and the emotional expressiveness of family members in relation to children's understanding of emotions (Cassidy et al., 1992). In essence, even when parents do not overtly display their emotions during family interactions, the home environment significantly influences children's ability to comprehend and communicate emotions, as documented by Halberstadt et al. (1995). Similarly, it has been observed that explicit conversations about various emotions (Huttenlocher, 1998), as well as high‐quality discussions about different emotions between parents and toddlers (Brownell, 2013), predict children's prosocial behaviours, which have previously been linked to children's emotion understanding ability (Denham, 1986). However, children in institutional care often lack personalized, emotionally meaningful interactions (Dozier et al., 2012). These interactions tend to be more transactional and less emotionally engaging. Additionally, institutionalized children engage in fewer social interactions with caregivers and peers, which are essential for learning about emotions. Furthermore, caregivers in many low‐resource countries may not have received adequate training in fostering emotional development among these children or may not prioritize it due to the overwhelming demand to meet the basic needs of the children under their care (Whetten et al., 2009). In the absence of explicit labelling and discussion of emotions, these children miss out on crucial learning opportunities. In the given context, the findings from the informal discussion with the caregivers appear to align with the existing literature mentioned above. Therefore, it could be inferred that the deprivation of an early socio‐emotional environment, limited interaction with caregivers and lack of meaningful caregiver–child interactions might be the primary factors contributing to the institutionalized children's underdeveloped emotion identification skills.

On the other hand, the emotion situation task was relatively more challenging than the emotion identification task, as it necessitated children to infer emotions from vignettes depicted by picture cards. The descriptive analysis of the emotion situation task (refer to Figure 2) further revealed that institutionalized children demonstrated a better understanding of sadness, followed by happiness, anger and fear. However, the discrepancy could be explained by the fact that as sadness had fewer vignettes compared to other emotions, this might have allowed institutionalized children to perform better in that particular domain. Furthermore, institutionalized children provided fewer confused responses in the emotion situation task compared to the emotion identification task. Nevertheless, an upward trend was observed when assessing the percentage of incorrect responses related to anger and fear, suggesting that these children faced greater difficulty comprehending anger and fear situations compared to sadness and happiness. These findings contradict those of Fries and Pollak (2004), who discovered that post‐institutionalized children struggled with both emotion identification and situation tasks and experienced difficulties matching emotional expressions to situations involving happiness, sadness and fear, although their performance in recognizing anger‐related situations was similar to their peers. Moreover, previous studies have reported that children within this age group tend to confuse anger and sadness when confronted with identical scenario tasks and frequently misinterpret these two emotions (Levine, 1995). Similarly, Fries and Pollak (2004) found that community children often selected sadness when faced with anger‐inducing situations, whereas post‐institutionalized children did not prefer sadness over anger. In contrast, in our study, a small number of institutionalized children did express a tendency to choose sadness instead of anger in situations involving anger. The cultural context in India may explain this observed phenomenon, where children are encouraged to refrain from exhibiting socially disengaging emotions, as such displays could pose a potential threat to positive group dynamics and interpersonal relationships. In contrast, socially engaging emotions such as sadness or fear are viewed as less disruptive as they tend to elicit support from others (Kathuria et al., 2022). Along the same line, research by Raval and Martini (2009) on middle‐class mothers in urban India indicated that these mothers considered their children's expressions of anger to be less acceptable than expressions of sadness or physical pain. Moreover, they reported being more likely to comfort and encourage expressions of sadness or pain than of anger. As a result, school‐age children from these families expressed anger less frequently in hypothetical situations involving their mothers compared to sadness and pain (Raval et al., 2007).

From the above discussions, it could be concluded that institutionalized children responded more accurately in the emotion situation task than in the emotion identification task (refer to Figure 4). The qualitative analysis of the video recordings could provide an explanation for this. The contextual cues in the emotion situation task in the form of vignettes helped the institutionalized children to perform better than the emotion identification task. However, due to the qualitative nature of our study and the small sample size, we were constrained to do any additional statistical analysis beyond descriptive statistics to examine whether the differences between the two tasks were statistically significant.

Lastly, the findings of the pilot study on the emotion recognition ability of non‐institutionalized children revealed a high level of accuracy in identifying emotions. Specifically, the children exhibited 100% accuracy in recognizing happiness, followed by sadness, anger and fear. Furthermore, they achieved over 95% accuracy in identifying emotions of sadness, anger and fear (See Figure 1). Regarding the emotion situation task, the non‐institutionalized children demonstrated 100% accuracy in recognizing happiness, followed by sadness, fear and anger. They also achieved over 95% accuracy in situations involving sadness and fear (See Figure 2). However, the children raised in a family environment responded more incorrectly to scenarios involving anger than their institutionalized peers (See Figure 2). This finding aligns with previous research conducted by Fries and Pollak (2004), which indicated that post‐institutionalized children were more likely to select anger when presented with anger‐provoking situations, while non‐institutionalized children tended to choose sadness over anger in similar scenarios. Although statistical analysis could not be performed due to the limited sample size and qualitative nature of the study, we analysed the performance of both institutionalized and non‐institutionalized children by using descriptive analysis in terms of the percentage of accurate responses. Moreover, the qualitative analysis from the detailed video recordings revealed that non‐institutionalized children responded more confidently, requiring minimal prompting and less time in comparison to institutionalized children. Also, non‐institutionalized children could easily label, identify and understand emotions without needing four alternatives for each item.

Limitations

Due to access issues and time constraints, data were collected from a small sample of institutionalized children. The children were selected from three different CCIs in urban areas, so the sample is not representative. It was time‐consuming because the data collection and analysis methods were primarily qualitative. Moreover, using an unequal number of items in the tool for certain emotions (fear in the emotion identification task and sadness in the emotion situation task) might potentially introduce bias or impose limitations on the findings. Also, the current study's findings cannot be generalized because different sociodemographic, psychosocial and institutional factors vary and may impact the results.

Another limitation of the present study pertains to the nature of the task employed. As both tasks required verbal responses, it may not be suitable for children who comprehend emotions but are unable to articulate their responses verbally. Although the drawings provided participants with visual representations of characters' facial expressions, enabling them to infer emotions based on these cues, the necessity for verbal responses poses a challenge for children with language‐related difficulties. As a result, the task was not entirely designed to be non‐verbal, further limiting its applicability to a broader range of participants.

Strengths

Despite the limitations, the present study exhibits several methodological advantages, strengthening its validity. Firstly, the data acquisition method employed in this study incorporated a child‐friendly approach by introducing visual materials such as cartoon characters and picture cards. Consequently, the children found both tasks more enjoyable, resulting in increased enthusiasm and engagement. Secondly, a rigorous peer review process, content validity assessment and meticulous translation–back translation and inter‐rater reliability were conducted to enhance the instrument's psychometric properties, ensuring the design, cartoon characters and picture cards for both tasks were of high quality. Moreover, aligning the task scenarios with the children's everyday experiences established a stronger connection, allowing for more accurate responses. Additionally, the emotion‐neutral depiction of characters in the picture cards added a degree of challenge, preventing any ceiling effects. Furthermore, the equal representation of male and female characters in both tasks minimized the likelihood of stimulus‐specific effects on children's performance. Lastly, data were collected over two separate sessions to accommodate the short attention spans of young children while ensuring the tasks' item count was kept to a minimum to prevent exhaustion.

CONCLUSION

The present study is significant from a psychosocial standpoint. A thorough evaluation was imperative to understand the institutionalized children's performance across both tasks and to identify potential areas of concern. By assessing the emotion understanding abilities of these children, valuable insights can be gleaned to aid researchers and stakeholders in designing and implementing suitable intervention strategies. Nevertheless, further research is necessary to understand these children's emotion comprehension skills that support and promote their overall well‐being, particularly in low‐resource and developing nations and across various forms of institutional care.

To achieve the desired outcome, it is crucial to raise awareness among caregivers and management authorities regarding the importance of emotion understanding in institutionalized children's lives. Furthermore, equipping the caregivers with specialized and extensive training to facilitate an enriching emotional learning experience for institutionalized children could be a progressive path forward. Additionally, the caregivers could be trained to utilize various practical strategies to incorporate emotion labelling and conversation into daily activities, thereby helping to bridge this gap. Providing more individualized attention, where caregivers can model and reinforce the use of emotion labels during one‐on‐one interactions, might also be beneficial. Also, integrating emotional and social learning (ESL) programmes into the routine activities of children could assist institutionalized children by providing different avenues to express, label and understand their emotions appropriately. Considering these factors, policymakers can improve the infrastructure of CCIs by creating more stimulating environments and recruiting adequately skilled personnel to cater to the unique needs of these children.

AUTHOR CONTRIBUTIONS

Manaswini Mishra: Conceptualization; investigation; writing – original draft; methodology; writing – review and editing; formal analysis. Rooplekha Khuntia: Conceptualization; methodology; writing – review and editing; supervision.

CONFLICT OF INTEREST STATEMENT

The authors have no competing interests, financial or non‐financial.

ETHICAL APPROVAL

The present work has been approved by the Institutional Ethics Committee for Human Research (NISER/IEC/2022‐06), National Institute of Science Education and Research.

ANALYTIC CODE

The analytic code necessary to reproduce the analyses presented in this paper is available from the corresponding author.

MATERIALS

The materials necessary to attempt to replicate the findings presented here are available from the corresponding author.

PREREGISTRATION

The analyses presented here were not preregistered.

Supporting information

Data S1.

Mishra, M. , & Khuntia, R. (2025). Emotion understanding among institutionalized preschool children in India: A visual‐based approach. British Journal of Developmental Psychology, 43, 139–160. 10.1111/bjdp.12530

DATA AVAILABILITY STATEMENT

The raw data (video recordings) necessary to reproduce the analysis presented here are not publicly accessible because of ethical considerations.