Self-concept refers broadly to the internal representation of feelings and thoughts that a person has about themselves (Campbell et al., 1996). Self-concept is continuously shaped through the interaction between internal factors, such as biological and cognitive, and external factors such as the social environment, culture, and life events (Jetten et al., 2011). As a result, individuals construct their sense of self through the interplay between their identity, which identifies them as unique and different from others, and social identities based on their perceived group memberships (Tajfel and Turner, 1986). Research shows that self-concept is quite stable throughout a person’s life (Church et al., 2012); however, this stability might be disrupted by significant life events (Markus and Kunda, 1986) such as neurological injuries, major life transitions, social evaluation, and interventions targeting self-schema (Beadle et al., 2016).

A traumatic brain injury (TBI), formally defined “as an alteration in brain function, or other evidence of brain pathology, caused by an external force,” Menon et al. (2010, 1638) is a major event that might significantly destabilize a person’s life. TBI is a worldwide critical health issue with an annual incidence of 69 million persons globally, and a growing number of cases every year (Dewan et al., 2019). For those who survive, recovery depends on several factors, most importantly the severity of the injury, and can take from weeks to years. A 10-year retrospective study showed that, for the majority of cases, attaining a significant recovery could take up to 6 years (Huang et al., 2010).

A person’s life after TBI presents with many challenges stemming from physical, cognitive, behavioral, and emotional changes post-injury (Corrigan and Hammond, 2013). The associated deficits that result from this kind of injury, affecting cognitive (e.g., poor concentration, memory difficulties), physical (e.g., headache, dizziness), emotional (e.g., irritability, depression) or motor skills, often negatively impact the everyday functioning of the individual (Lundin et al., 2006). Consequently, it is common to experience psychological distress over time (Pagulayan et al., 2008).

The majority of people with TBI report symptoms of depression at least once, and significantly more times than people that have not suffered this kind of injury (Jorge et al., 2004; Hart et al., 2012). Furthermore, people with TBI and major depressive disorder are more susceptible to present comorbid anxiety (Jorge et al., 2004; Bombardier et al., 2010) as well as a lower quality of life (Bombardier et al., 2010). Moreover, people with TBI are prone to experience aggressive behaviors, suicidal ideation (Tsaousides et al., 2011; Mackelprang et al., 2014), and lower life satisfaction (Juengst et al., 2015). Social support is also affected. Persons with TBI report poor satisfaction with social support and lower locus of control in comparison with persons without TBI (Izaute et al., 2008). Lastly, emotional distress in persons with TBI has been associated with changes in self-concept (Beadle et al., 2016).

Changes in self-concept in this population have been studied using instruments that try to capture pre-post injury characteristics regarding self-concept, self-esteem, trauma, and personality (Tyerman and Humphrey, 1984; Nochi, 1998; Cantor et al., 2005; Rush et al., 2006; Cloute et al., 2008; Gracey et al., 2008; Muenchberger et al., 2008; Carroll and Coetzer, 2011; Gelech and Desjardins, 2011; Jones et al., 2011; Douglas, 2013; Downing et al., 2013; Lapadatu, 2015; Reddy et al., 2015; Riley and Hagger, 2015). All these studies show that pre-injury self-concept is more positive than post-injury self-concept. Overall, patients with TBI reported significant self-concept change, psychological distress (anxiety and depression) related to the change, strong correlations between affective distress and self-discrepancies, sexual changes related to fatigue, low confidence, pain, decreased mobility, feelings of being unattractive, construction of TBI as a deficit, loss of self in terms of unclear self-knowledge about their characteristics and capabilities and feelings of shame about the injury, among others (Tyerman and Humphrey, 1984; Nochi, 1998; Cantor et al., 2005; Rush et al., 2006; Cloute et al., 2008; Gracey et al., 2008; Muenchberger et al., 2008; Carroll and Coetzer, 2011; Gelech and Desjardins, 2011; Douglas, 2013; Downing et al., 2013; Reddy et al., 2015). Additionally, some studies indicate that people who have acquired a TBI have shown the need to grieve their previous selves in order to reconstruct their current selves (Nochi, 1998; Gelech and Desjardins, 2011). Cognitive inflexibility has also been proposed as a mechanism that sustains psychological symptoms after a traumatic brain injury, including poor adjustment to changes post-injury (Faulkner et al., 2020). Inflexibility could affect a person’s response to their injury and lead to poor outcomes (Broshek et al., 2015).

Changes in self-concept have been studied using several different methods, one of which is the repertory grid technique (RGT). This method was developed by Kelly (1955) as a tool to understand the personal constructs that every individual uses, and it was originally known as The Rep Test (Feixas and Cornejo, 1996; Fransella et al., 2004). For Kelly, personal constructs are the basic units of knowledge and analysis —not isolated but interconnected in a hierarchical network— which allow the individual to make predictions about themselves and their surroundings. Personal constructs are bipolar discriminations that can be described through verbal tags, such as happy-sad or captivating-boring, and can also operate as pre-verbal constructions to make life experiences more predictable (Bannister, 2003). The RGT has several methodological variants in form and administration.

Gracey et al. (2008) used a variation of the RGT to explore the construction of meanings and self-concept of 32 individuals with a brain injury through a group exercise that allowed eliciting the constructs by comparing pre-injury, current and ideal selves. The elicited constructs were then classified using thematic analysis. Most of the constructs exhibited themes related to the individual’s experience of the self in the world and their functioning in social and practical activity. That is the only study that has employed the RGT with individuals who have acquired a TBI; however, it only considered it as a qualitative method while the repertory grid could also provide significant quantitative information. Moreover, Gracey et al. (2008) applied the technique in a group scenario, while to understand the personal meanings it will be useful to apply it in an individual scenario.

Research about self-concept after TBI is limited in comparison to research looking at changes in cognition and general functioning (Bryson-Campbell et al., 2013; Manley et al., 2017; McInnes et al., 2017; Tsai et al., 2021; Villa et al., 2021). The current study aims to contribute to this body of knowledge by exploring ways of self-construing post-injury. Specifically, this study will explore: (a) whether distances between pre- and post-injury self-concept relate to mental health, and (b) whether a polarized view of self relates to mental health. The RGT is particularly suited for this endeavor due to its theoretical and methodological underpinnings explained above. Our study aims to use the RGT as a source of quantitative information and present it as a tool to assess the distances between individuals’ views of their present, past, and ideal selves, as well as polarization of self-concept, and their possible relations with depression, anxiety, satisfaction with life, and quality of life 6 months after experiencing a TBI. Given ample evidence demonstrating the relationship between self-concept discrepancy, polarization, and psychological symptoms, we expect to find a similar relationship in this population (Feixas et al., 2008, 2021). Due to the novelty of the use of the RGT for this purpose and with people with a TBI, the present work is proposed as an exploratory study.

The Research Bioethics Committee of the Eugenio Espejo Hospital approved the protocol of the present study. This research study was performed in accordance with the principles stated in the Declaration of Helsinki. The data presented here were collected as part of a larger longitudinal study tracking recovery from TBI. The data for the larger longitudinal study were collected between January 2018 and March 2020 at three assessment points: hospitalization, six (±2) months post-injury, and 12 (±2) months after the injury. For this study, we report information from participants assessed six (±2) months post-injury.

Six months after the injury, the mean depression score was 7.24 [5.42, 9.12] (SD = 5.40). Twenty-three (70%) patients presented mild depression and 10 (30%) presented moderate-to-severe depression. The mean anxiety score was 5.54 [4.12, 6.96] (SD = 4.14) Twenty-seven (82%) patients showed mild anxiety and six (18%) moderate-to-severe anxiety.

Regarding satisfaction with life 6 months after the injury, six (18%) participants were dissatisfied, four (12%) were slightly dissatisfied, four (12%) were slightly satisfied, 15 (45%) showed high satisfaction and four (12%) patients were extremely satisfied. Mean satisfaction with life was 23.30 [20.72, 25.54] (SD = 7.17).

Concerning quality of life 6 months after the injury, considering the proposed cutoff (<60) for identifying low or impaired HRQoL with the QOLIBRI (Truelle et al., 2010; Wilson et al., 2017), 12 (36%) participants showed low or impaired HRQoL. The mean quality of life score was 61.19 [54.87, 66.97] (SD = 16.39).

The mean number of constructs elicited in the RGT was 19 (SD = 3.90) in total, 8 (SD = 3.57) self-generated, and 11 (SD = 4.31) elicited from the list. The Euclidean distance between present-self and self-before the lesion was 0.39 [0.31, 0.47], between present-self and ideal-self was 0.35 [0.29, 0.43] and between the self before the lesion and the ideal self was 0.37 [0.31, 0.43]. As the CIs of all these distances overlap, there is no evidence of significant differences between them.

Table 1 shows the correlations between all the self-reported measures. The PHQ-9 shows a significant moderate negative correlation with the SWLS and the QOLIBRI, and a significant high positive correlation with the GAD-7. That is, the higher the depression score is, the lower the satisfaction with life and quality of life scores, and the higher the depression score is, the higher is the score on the anxiety measure. We also found significant correlations of the GAD-7 with the SWLS as well as with the QOLIBRI, both low and negative.

Table 2 shows correlations for each score of the self-reported measures against the Euclidean distances between all the possible pairs of elements. The distance between present self and the self before the lesion showed significant correlations with the scores on all measures except the GAD-7 scores: a low positive correlation with the PHQ-9, and low negative correlations with the SWLS, and the QOLIBRI. The distance between the present self and the ideal self also showed significant correlations with the scores on all measures except the GAD-7 scores: a low positive correlation with the PHQ-9 score, a low negative correlation with the SWLS and a moderate negative correlation with the QOLIBRI.

Also, we calculated the proportion of opposed pole construct ratings by pairs of elements. Table 3 shows these proportions and their correlations with all the self-reported measures. For the pair present self and self before the lesion we found significant correlations with the PHQ-9 (low positive), and with the SWLS and the QOLIBRI (moderate negative). The pair present self and ideal-self showed significant correlations with all the self-reported measures: moderate positive with the PHQ-9, low positive with the GAD-7, moderate negative with the SWLS and high negative with the QOLIBRI. The pair self before the lesion and ideal self, showed a significant low negative correlation with the SWLS.

Finally, we calculated the proportion of polarization by element and for all the elements. The mean proportion for each element, the 95% CI and the correlations against all the self-reported measures scores are presented in Table 4. Significant correlations were found between present self-polarization and the PHQ-9 and GAD-7 scores, both correlations being low negative. Moreover, there was a low negative correlation between total polarization and the GAD-7 score.

In this study, the RGT was used as a source of quantitative information to understand self-concept after a TBI and its possible relations with depression, anxiety, satisfaction with life, and quality of life measures. This adapted version of the technique seems adequate to obtain the necessary information to identify perceived distances between aspects in the personal construction of the self in this specific population.

In this study, 70% of the patients presented mild depression and 30% presented moderate-to-severe depression, 82% of the patients showed mild anxiety and 18% moderate-to-severe anxiety. Analysis of the element distance measure indicated no significant differences in how positively (in terms of proximity to the ideal self) patients view themselves after the lesion compared to how they have been before the lesion occurred. However, we found that the distance between present self and the ideal self showed a significant and positive association with the score representing symptoms of depression, and negative associations with the satisfaction with life and with the quality of life. Of note, a less polarized view of self-construing was associated with higher levels of depression and anxiety in the study participants. This result is not consistent with findings in other groups in which polarization and inflexibility are typically associated with lower levels of well-being (Feixas et al., 2008, 2021; Aguilera et al., 2019; Faulkner et al., 2020). It is possible that, 6 months post-injury, a less well-defined sense of self can result in greater distress.

Levels of congruence or incongruence in several kinds of self-representation have been linked to emotional problems such as depression, consistent with the results of the current study (Higgins et al., 1986; Higgins, 1987). Discrepancy in aspects of self-concept, such as differences between present- and ideal-self, has been well documented in other populations (Feixas et al., 2008, 2021). Consistent with this, our findings suggest that viewing the current self as dissimilar to both the pre-lesion self and the ideal self is associated with affective distress in people who have suffered a TBI and can be compared to those from Cantor et al. (2005) who examined the utility of self-discrepancy theory in relation to depression and anxiety post-TBI and found strong correlations between anxiety, depression, and self-discrepancies. However, we only found positive correlations with depression, and negative correlations with satisfaction with life and quality of life. These correlations between emotional distress and self-discrepancies may be related to people who have suffered a TBI perceiving their past and present selves in a pessimistic manner, as Shields et al. (2016) have pointed out.

Based on our results, the RGT can play an important role in TBI research, by exploring self-concept and correlates to emotional outcomes. However, it may also provide a platform to formulate treatment approaches. An increase in awareness of changes post-injury has been tied to better outcomes in neuropsychological treatments (Sherer et al., 1998; Hurst et al., 2018). Thus, clarifying the areas in which there is a distance between current and ideal self can itself be therapeutic by increasing awareness of deficits after TBI, thereby opening the door to address them in the neurorehabilitation process. Furthermore, exploring and resolving distances between pre- and post-injury self-concept could further a person’s adjustment to disability and improve mental health in the process. In addition, clarifying self-concept after injury may lead to a more well-defined sense of self and lessened psychological distress. Holistic approaches to neurorehabilitation indeed include both the recognition of changes in the pursuit of higher functioning as well as the construction of personal meaning within the new matrix of abilities, relationships, and expectations brought about by the traumatic brain injury (Ben-Yishay et al., 1985).

This study has limitations regarding sample size and gender representativeness. Further research using the RGT with a bigger sample of patients and more women participants that have suffered a TBI could lend further support to the notion that discrepancies and/or polarization of self-concept occur in this population and affect their mental health.

This is the first study to use quantitative analysis of the RGT to explore the relationship between aspects of self-construing and mental health in sufferers of traumatic brain injury, 6 months post-lesion. Mental health providers should consider exploring self-concept discrepancies and polarization when working with this population.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Research Bioethics Committee of the Eugenio Espejo Hospital. The patients/participants provided their written informed consent to participate in this study.

GM: study design, supervision, and paper writing. VC: paper writing. JV-O: data collection, analysis, and paper writing. AR-L: study design and paper writing. JA-L: paper writing. CP: study design and paper writing. All authors contributed to the article and approved the submitted version.

This work was supported by the (Dirección General de Investigación y Vinculación, Universidad de Las Américas, Quito, Ecuador) under grant (PSI.GM.1705).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.