Sensory processing sensitivity and its associations with guilt, shame, self-esteem, and neuroticism

Abstract

Background

Sensory Processing Sensitivity (SPS) is a trait linked to deeper processing of stimuli and heightened emotional reactivity. These characteristics suggest a potential link to more intense self-conscious emotions. The objective of this study was to investigate the associations between SPS, feelings of guilt, shame, and self-esteem, and to test whether these relationships are independent of the influence of neuroticism.

Methods

We conducted a cross-sectional study using data from an online survey of Czech adults (n = 1012; 49.3 ± 16.7 years, 50.4% female). Participants completed measures of SPS (Sensory Processing Sensitivity Questionnaire, SPSQ; Highly Sensitive Person Scale), feelings of guilt and shame (Guilt and Shame Experience Scale), self-esteem (Rosenberg Self-Esteem Scale), and neuroticism (the neuroticism subscale of the Big Five Inventory). The associations were examined using linear and logistic regression models, with adjustments for neuroticism and key demographic variables.

Results

Linear regression analyses showed that higher SPS was significantly associated with increased feelings of guilt and shame, and with lower self-esteem. After adjustment for neuroticism, the association between SPS and self-esteem was no longer significant (β ≈ 0.03, p > 0.05), whereas the β-coefficients for feelings of guilt and shame were reduced but remained significant. Logistic regression analyses comparing low, medium, and high SPS groups and, separately, the equivalent levels on the Sensory Sensitivity subscale of the SPSQ, indicated that highly sensitive individuals were more likely to report feelings of guilt alone and combined guilt and shame, with odds ratios (ORs) ranging from 4.42 (95% CI 2.17–8.99, p < 0.001) to 6.38 (95% CI 3.08–13.25, p < 0.001). No significant associations emerged between SPS and feelings of shame alone or low self-esteem. Analyses using alternative SPS measures yielded broadly similar results, with the Highly Sensitive Person Scale showing even stronger associations with feelings of guilt and shame, while again no effect was found for self-esteem.

Conclusions

Highly sensitive individuals appear to be more prone to experiencing heightened feelings of guilt and, to a lesser degree, shame. However, the initially observed negative association between SPS and self-esteem was no longer significant after neuroticism was included in the model.

Introduction

Guilt and shame, while closely related, represent distinct self-conscious concepts with different psychological implications [1, 2]. Guilt is a complex psychological construct that combines a sense of responsibility for one’s actions with a negative self-assessment of behaviours perceived as violating personal or societal moral standards [1]. It typically arises when an individual’s behaviour deviates from internalised moral norms [3]. Current research further distinguishes between reparative guilt, which focuses on correcting a specific wrongdoing, and interpersonal guilt, which results from a perceived failure to fulfil relational obligations or causing harm to others [4, 5]. Guilt can serve dual roles. It can act as a powerful trigger of negative emotions, darkening one’s mental state [6]. When this negative experience becomes chronic and pervasive, as is often the case with interpersonal guilt, it can be maladaptive and linked to psychological distress and conditions such as depression and anxiety [7–9]. Chronic guilt is particularly problematic when accompanied by rumination, a repetitive cognitive process that maintains and increases feelings of guilt over time [10]. Conversely, guilt can also catalyse personal growth by motivating individuals to correct their missteps and reach their potential. This adaptive, motivational function is characteristic of reparative guilt, which is considered a prosocial emotion that fosters empathy and corrective behaviour [5, 6]. Thus, feelings of guilt that are appropriate to the situation are essential for moral motivation because they move individuals to maintain or restore interpersonal relationships and cooperate with others [11]. Therefore, feelings of guilt are characterised by a focus on specific actions and can lead to constructive self-improvement and adaptive behavioural adjustments [12].

Shame, by contrast, involves a more global negative evaluation of the self and a sense of personal failure [13]. It typically arises in response to violations of social norms and values, occurring when individuals feel they have fallen short of their ideal self [14]. Traditionally, the social-adaptive perspective has framed shame as a uniformly maladaptive emotion due to its strong links to withdrawal, avoidance, and psychopathology [5, 15]. However, more recent functionalist approaches argue that shame can also have adaptive functions in certain contexts. It can lead us to try to repair the damage, adapt to social expectations, and restore relationships that have been damaged by our actions [2, 16]. Nevertheless, the maladaptive potential of shame remains significant, especially when it becomes a chronic personality trait or when one experiences it in situations where there is little chance of changing anything [16, 17]. In these cases, shame leads to deep feelings of worthlessness, pervasive self-blame [18], and can significantly impair relationships with both oneself and others [19]. However, while feelings of guilt are associated with taking responsibility for one’s actions, feelings of shame undermine self-esteem and focus more on the overall negative self-image and evaluation [2].

Self-esteem is an individual’s subjective evaluation of their own worth, comprising three components: perceived competence, sense of worthiness, and overall self-evaluation [20]. It is usually distinguished from self-concept, which comprises beliefs about one’s personal characteristics, attributes, and abilities. Whereas self-concept stems from cognitive links between the self and specific attributes, self-esteem emerges when those links are evaluated with positive or negative emotion [21, 22].

Importantly, both feelings of guilt and shame have been linked to poorer mental health outcomes, including depression, anxiety, social isolation, and suicidal ideation [7, 23, 24], particularly under conditions of psychological distress [25]. Furthermore, low self-esteem has also been associated with mental health difficulties [26], as well as across various life domains such as relationships, feelings of happiness, and professional achievement [27]. Therefore, recognising the factors that contribute to elevated feelings of guilt, shame, and low self-esteem is crucial for comprehensive understanding and the development of targeted interventions in mental health care. Among these factors, the impact of personality traits [28–31], parenting styles [5, 9, 32, 33], and other environmental influences [34] has been somewhat explored. However, one factor that remains under-researched is sensory processing sensitivity (SPS), which may influence the intensity of self-evaluative emotions in ways that are not yet fully understood.

SPS is a personality construct first formally operationalised by Elaine and Arthur Aron in 1997 and defined as a personal predisposition to process subtle stimuli more deeply than others [35]. This neurobiological trait is characterised by increased brain activity in areas focused on cognitive information processing, memory, and attention [36], as well as emotional reactivity [37]. Neuroimaging studies have revealed that highly sensitive persons (HSPs) show increased activity in areas of the brain associated with empathy and self-related processing when viewing emotional stimuli [38]. Moreover, they process all internal and external sensory input, including social stimuli, more intensively [39]. Importantly, SPS is embedded within the broader theoretical framework of environmental sensitivity [40], which brings together several interrelated models to explain differences in individuals’ responses to their environment. Among these, the theory of Differential Susceptibility argues that some individuals respond more strongly than others to both adverse and favourable environments [41]. Similarly, Biological Sensitivity to Context emphasises individual differences in physiological stress reactivity (e.g., heart rate, cortisol response), calibrated by early-life experiences [42]. Taken together, these theories explain why some individuals are more responsive to environmental influences than others and suggest that individuals with heightened reactivity may be especially vulnerable in stressful environments but also derive greater benefit from supportive ones [42].

Based on its theoretical underpinnings, SPS is highly relevant to both personality psychology and mental health because it provides significant benefits while also presenting potential vulnerabilities. For instance, SPS brings advantages such as improved perception of subtle environmental nuances [43], enhanced creativity and cognitive processing [44], and increased empathy and care for others [38]. However, at the same time, this heightened sensitivity can lead to sensory overload from loud noises or bright lighting [45], necessitating lifestyle adaptations to minimise overstimulation [46, 47]. SPS is also associated with increased emotional reactivity and cognitive strain under stress and challenging environments [48, 49] and with a greater likelihood of negative psychological outcomes such as depression, anxiety, stress [35, 50] or neuroticism [51].

Heightened emotional reactivity and deeper processing of internal and social stimuli in HSPs may predispose them to experience moral and self-evaluative emotions such as guilt, shame, and self-esteem more intensively [37, 52]. There are several ways to explain this relationship. First, HSPs, due to their heightened empathy, may feel guilty even for things they are not directly responsible for. Simply seeing someone suffering appears to be sufficient [38, 53]. Second, because they process moral information more deeply, they may evaluate their actions and their consequences more intensely, leading to stronger moral emotions [49]. Third, their tendency to think about things for a long time can turn ordinary situational guilt into a chronic, harmful problem [54, 55]. Regarding shame specifically, HSPs may be more prone to experience shame due to their heightened sensitivity to social cues and the judgments of others, as they are more likely to detect subtle signals of social disapproval or rejection [35]. Furthermore, the intense emotional experiences typical of HSPs may be perceived as excessive or inappropriate in certain social contexts, which can cause shame about their own emotional reactivity [56]. Overall, as environmental sensitivity frameworks suggest that HSPs are differentially susceptible to negative contexts [41], adverse experiences may also elicit more intense feelings of guilt or shame compared to less sensitive individuals [57].

Recent research indicates that individuals with higher levels of SPS tend to exhibit lower self-esteem [37, 58, 59]. HSPs show greater sensitivity to negative social feedback, tend to internalise criticism more easily, and often experience feelings of being different or misunderstood in less accommodating environments [60]. Furthermore, the tendency to experience self-conscious emotions more intensely can create a problematic pattern in which shame and low self-esteem reinforce each other and increase susceptibility to maladaptive guilt [56].

However, it is important to note that SPS has often been confounded with neuroticism—another trait associated with emotional instability and vulnerability to distress [35]. Although the two constructs are empirically correlated, they are theoretically and psychometrically distinct [35, 61]. Therefore, controlling for neuroticism is essential when examining associations between SPS and psychological outcomes. Several previous studies have adopted this approach, statistically adjusting for neuroticism to isolate the unique effects of SPS [56, 62]. The relationship between SPS, self-conscious emotions, and self-esteem may be further complicated by sociodemographic factors that independently influence these constructs. Research shows significant gender differences across all these domains. Women are more likely than men to exhibit high SPS [63], and they also report significantly higher levels of shame and guilt in various situations [33, 64, 65]. Conversely, men consistently report higher self-esteem than women across cultures [66]. Age-related differences have also been observed. Some aspects of SPS, such as low sensory threshold and ease of excitation, decrease linearly with age, whereas other aspects, such as aesthetic sensitivity, increase [67]. Similarly, guilt tends to rise from adolescence into old age, while shame decreases until middle adulthood and then increases again in later life [68]. Self-esteem, in turn, shows age-related increases from late adolescence to middle adulthood [66]. These findings highlight the importance of accounting for sociodemographic factors in integrated models.

In sum, by integrating SPS with guilt, shame, and self-esteem within an environmental sensitivity framework, this study addresses a significant theoretical and empirical gap. It explores whether SPS uniquely predicts moral emotions and self-evaluative outcomes, or whether these associations are better accounted for by underlying neuroticism. Importantly, our analyses consider sociodemographic characteristics, which allow for a more robust assessment of the unique role of SPS. To our knowledge, this integrated research has not yet been empirically conducted, representing a new contribution to the literature on SPS and self-related emotional processes.

Methods

Participants and procedure

We employed a questionnaire-based approach to collect the data for this study. An online survey was conducted in May and June 2020 among the Czech population aged 18 to 86 years. The data were collected by a specialised agency (Czech National Panel, Prague, Czech Republic) to achieve a sample balanced for age and gender in line with national demographics. Specific exclusion criteria were implemented to ensure data reliability. First, individuals who completed the survey in under ten minutes (despite it taking approximately one hour) were excluded, as it was unlikely they had provided thoughtful responses. Second, respondents who provided inconsistent answers to repeated control questions on age, weight, and height (i.e., reporting a difference of two or more units of measure) were also excluded. Based on these criteria, we excluded 120 subjects. Furthermore, 128 respondents who did not provide any responses to the SPSQ items were also removed from the analysis. The final sample comprised 1012 respondents (49.3 ± 16.7 years, 50.4% female).

At the beginning of the survey, all participants were required to provide explicit consent after receiving written information about the study’s aim and the anonymised, confidential handling of data. They were then introduced to the online survey system. Given the online nature of the survey, electronic informed consent was obtained. Participation in the survey was voluntary, and respondents could withdraw at any point without providing a reason. The study design received approval from the Ethics Committee of the Faculty of Theology at Palacký University in Olomouc (No. 2020/06).

Measures

We used two instruments to assess SPS: The Sensory Processing Sensitivity Questionnaire (SPSQ) and The Highly Sensitive Person Scale (HSPS).

The SPSQ is an instrument developed and validated in the Czech environment by Malinakova et al. [69]. The initial query was phrased as follows: “Please indicate to what extent you think that, compared to other people, you are sensitive to the following stimuli, conditions, or experiences.” Each item was rated on an eleven-point scale, ranging from “Compared to others, I am not sensitive to them at all” (0), through “About the same as the people around me” (5), to “Much more sensitive than the people around me” (10). Following the initial question, participants responded to eight sensory sensitivity items (light, sounds, smells, taste, tactile stimuli such as touch and clothing, hunger, heat, and cold) and eight other sensitivity items (one’s own emotions, others’ emotions, sudden changes, one’s inner world, the need to do many things at once, criticism, the need for harmony in life, and the need to make decisions). Total scores ranged from 0 to 160, with higher SPSQ scores indicating greater SPS. We also employed the Sensory Sensitivity subscale of the SPSQ questionnaire, comprising only the eight sensory sensitivity items. In our sample, Cronbach’s α was 0.93 for the full SPSQ and 0.90 for the SPSQ-S subscale.

The HSPS, developed by Aron and Aron [35], comprises 27 items assessing highly sensitive person characteristics (e.g., “Are you easily overwhelmed by strong sensory input?” or “Do you become unpleasantly aroused when a lot is going on around you?“). Responses are given on a seven-point Likert scale ranging from “strongly disagree” (1) to “strongly agree” (7). Higher HSPS scores indicate a greater degree of SPS. Cronbach’s α was 0.9 in our sample.

For statistical analysis, a consistent tertile-based categorisation was applied to the total scores of the SPSQ (range 0–160), its Sensory Sensitivity subscale SPSQ-S (range 0–80), and the HSPS (range 27–189), resulting in three groups: low, medium, and high sensitivity. This approach follows the recommendations of Lionetti et al. [70], who highlight its utility despite potential population variation, and is also supported by its use and robust findings in recent Czech research [71]. For regression models, we retained the continuous summary scores of both instruments.

Self-esteem was measured using the Czech version [72] of Rosenberg self-esteem scale (RSES) [73], a widely utilised self-report assessment. This scale comprises ten items rated on a four-point Likert scale ranging from “strongly agree” (1) to “strongly disagree” (4). Positively worded items were reverse-coded to compute the total RSES score. We then generated a reversed total score, where higher values indicated lower self-esteem. The scores were subsequently dichotomised, with a value of 1 indicating the presence of low self-esteem.

Guilt and shame experience was assessed using the Guilt and Shame Experience Scale (GSES) [74]. This self-report tool comprises eight items, for example, “When I make a mistake, I cannot stop thinking about it.” or “I feel the need to explain or apologise for my actions.” Participants rated the extent to which they experienced these states or thoughts on a four-point scale, ranging from “Not at all” (1) to “Significantly” (4). Responses were summed to obtain a total GSES score, with higher values indicating a greater level of feelings of guilt and shame. For statistical analyses, we primarily used the summary score of the GSES and its subscales, as this approach preserves the full range of individual differences. For logistic regression models, we dichotomised the scores at the scale’s midpoint. Since the total GSES score has a potential range from 8 to 32, the theoretical midpoint is 20. This cut-off corresponds to an average item response of 2.5, which falls exactly between the verbal anchors “slightly” (2) and “moderately” (3). While we acknowledge the general statistical costs of dichotomisation e.g., [75, 76], this approach was chosen for its clear interpretability and grounding in the structure of the measurement instrument itself [74], which is considered a valid reason for dichotomisation in specific contexts where meaningful cut-off points can be established [75]. Cronbach’s α was 0.88 in our sample.

Furthermore, we analysed two specific GSES subscales separately: GSES-G (guilt subscale) and GSES-S (shame subscale), each comprising four items. Responses were summed to generate a score for each subscale. In our sample, Cronbach’s α was 0.83 for the GSES-G and 0.79 for the GSES-S. Consistent with the procedure for the full GSES, we used both continuous scores and dichotomised scores for the subscales. The dichotomisation was based on the same rationale, using the theoretical midpoint of their 4–16 score range, which is 10.

Neuroticism was assessed using the neuroticism subscale (BFI-N) of the Big Five Inventory [77, 78], a widely used assessment of personality traits. The BFI-N comprises eight statements designed to measure the tendency to experience tension and feelings of anxiety, for example, “I consider myself someone who is depressed, sad.” or “I consider myself someone who stays calm in tense situations.” Participants rated each statement on a five-point Likert scale, ranging from “strongly disagree” (1) to “strongly agree” (5). Positively worded items were reverse-scored so that higher total BFI-N scores indicated a greater degree of neuroticism. The continuous summary score was used for all statistical analyses. In this study, we employed the Czech version of the Big Five Inventory, validated by Hřebíčková [79]. Cronbach’s α was 0.84 in our sample.

We gathered sociodemographic information from the questionnaire, including age, gender, education level, and economic status. Full item wordings and response options for all scales used in the study are available in Supplementary Material 1.

Statistical analyses

As the first step, we described the background characteristics of the sample. Second, based on the Shapiro-Wilk test, we rejected the assumption of normal data distribution (SPSQ total score: W = 0.921, p < 0.001; Sensory Subscale score of the SPSQ: W = 0.879, p < 0.001). We then performed linear regression analysis using SPS as the independent variable and guilt, shame, their composite score, and self-esteem as dependent variables. Each model was tested separately, first as a crude model, and consequently adjusted for age, gender, and education by including these covariates simultaneously with SPS in the same model; neuroticism was subsequently added in the next step as an additional covariate. Next, we repeated these analyses using the SPSQ-S and HSPS instead of the overall SPSQ score. Furthermore, as a sensitivity analysis, a logistic regression analysis was conducted with SPS (SPSQ, SPSQ-S, HSPS, each divided into three categories) as the independent variable and guilt, shame (GSES), and self-esteem (RSES) as dichotomised dependent variables. All logistic regression models were adjusted for age, gender, education, and neuroticism as in the linear regression models.

All analyses were performed using the statistical software package IBM SPSS version 27 (IBM Corp., Armonk, NY, USA).

Results

Description of the study sample

Table1 presents the sociodemographic characteristics of the sample. Of the whole sample, 14.5% of the respondents were classified as having low SPS. The majority, comprising 76.3% of the sample, fell into the medium SPS category. Additionally, 9.2% of the participants were classified as HSPs. The analysis revealed that SPS levels differed significantly by gender, education, and economic status (χ² tests, all p-values < 0.05 for gender and economic status, and < 0.01 for the highest education achieved). For example, the proportion of women in the high SPS group was substantially higher than that of men.

Table 1.

Sample characteristics by sensory processing sensitivity

	Number	%	Sensory Processing Sensitivity						P Value
			Low SPS		Medium SPS		High SPS
			Number	%	Number	%	Number	%
Gender									p < 0.05
Male	502	49.6	86	58.5	386	50.0	30	32.3
Female	510	50.4	61	41.5	386	50.0	63	67.7
Age									n.s.
18–34 years old	222	21.9	29	19.7	164	21.2	29	31.2
35–49 years old	275	27.2	44	29.9	210	27.2	21	22.6
50–65 years old	266	26.3	39	26.5	202	26.2	25	26.9
66–99 years old	249	24.6	35	23.8	196	25.4	18	19.4
Highest education achieved									p < 0.01
Elementary school	86	8.9	17	12.2	57	7.7	12	13.2
Secondary vocational school	367	37.9	67	48.2	260	35.2	40	44.0
Secondary school with graduation	354	36.6	43	30.9	284	38.5	27	29.7
College	161	16.6	12	8.6	137	18.6	12	13.2
Economic status									p < 0.05
Employee	465	48.1	73	52.5	356	48.3	36	39.6
Entrepreneur	62	6.4	4	2.9	55	7.5	3	3.3
In the householda/Unemployed	89	9.2	13	9.4	62	8.4	14	15.4
Student	51	5.3	6	4.3	35	4.7	10	11.0
Not working pensioner	300	31.0	43	30.9	229	31.1	28	30.8
Marital status									n.s.
Married/partnership	494	48.8	79	53.7	375	48.6	40	43.0
Single/divorced/widow(er)	518	51.2	68	46.3	397	51.4	53	57.0
Total	1012	100	147	14.5	772	76.3	93	9.2

^aIncluding maternity leave. Sensory Processing Sensitivity assessed by the Sensory Processing Sensitivity Questionnaire (SPSQ)

Sensory processing sensitivity, guilt, shame, and self-esteem

Table 2 presents the linear regression results for the associations of SPS with the feelings of guilt and shame, and low self-esteem. In the crude Model 1 (using the SPSQ total score), higher SPSQ scores were significantly associated with increased GSES total scores, feelings of guilt and shame, and lower self-esteem (standardised β = 0.17–0.36; p < 0.001). For a graphical illustration of the SPS–GSES relationship, see Fig. 1. After adjusting for demographic variables (age, gender, and education level), the associations remained statistically significant with only negligibly reduced effect sizes (Δβ ≤ 0.01); therefore, this model is described here but omitted from the table for brevity. In the fully adjusted model (controlling for demographics and neuroticism), the associations between SPSQ and both feelings of guilt and shame remained significant, whereas the association with self-esteem was no longer significant (β = 0.03; p > 0.05).

Table 2.

Associations of sensory processing sensitivity with guilt, shame, and self-esteem: results of linear regression, crude and fully adjusted for age, gender, education level, and neuroticism (standardised β and 95% confidence intervals)

			GSES total score			Guilt			Shame			Low Self-esteema
Model	Variable		B	β	95% CI	B	β	95% CI	B	β	95% CI	B	β	95% CI
1	SPSQ	Crude	0.07	0.34	0.06–0.08***	0.04	0.36	0.04–0.05***	0.03	0.26	0.02–0.04***	0.03	0.17	0.02–0.05***
		Adjusted	0.04	0.17	0.03–0.05***	0.02	0.20	0.02–0.03***	0.01	0.11	0.01–0.02***	−0.01	0.03	−0.02-0.01
2	SPSQ-S	Crude	0.08	0.22	0.06–0.10***	0.05	0.24	0.04–0.06***	0.03	0.17	0.02–0.04***	0.04	0.11	0.02–0.06**
		Adjusted	0.04	0.10	0.02–0.06***	0.03	0.12	0.01–0.04***	0.01	0.06	0.00–0.02*	−0.01	−0.02	−0.02-0.01
3	HSPS	Crude	0.11	0.47	0.09–0.12***	0.06	0.48	0.06–0.07***	0.04	0.38	0.04–0.05***	0.04	0.21	0.03–0.06**
		Adjusted	0.07	0.29	0.05–0.08***	0.04	0.31	0.03–0.05***	0.03	0.21	0.02–0.03***	−0.01	0.04	−0.02-0.00

SPSQ SPSQ total score, SPSQ-S Sensory Subscale score of the SPSQ, Guilt Guilt subscale of the GSES, Shame Shame subscale of the GSES

^aReversed summary RSES score: a higher score indicates lower self-esteem

* p < 0.05, ** p < 0.01, *** p < 0.001

Fig. 1.

Linear association of sensory processing sensitivity and guilt and shame. Notes: GSES_sum = GSES total score; SPSQ_sum = SPSQ total score

Furthermore, Model 2 (SPSQ-S subscale) demonstrated similar patterns. In the crude model, higher sensory sensitivity was significantly associated with the feelings of guilt and shame, and low self-esteem, with β coefficients ranging from 0.11 to 0.24 (p < 0.001). After adjusting for demographic variables, these associations remained significant, with only slight reductions in effect sizes (Δβ ≤ 0.02). In the fully adjusted model (including neuroticism), the associations between SPSQ-S and feelings of guilt and shame, although reduced, remained significant, whereas the link with self-esteem was no longer significant.

Model 3 (HSPS) showed a similar pattern of results. In the crude model, higher HSPS scores were significantly associated with elevated feelings of guilt and shame, and low self-esteem (β = 0.21–0.47; p < 0.001). Controlling for age, gender, and education had minimal impact on these associations. In the fully adjusted model (controlling for demographics and neuroticism), HSPS remained significantly associated with feelings of guilt (β = 0.31; p < 0.001) and shame (β = 0.21; p < 0.001), while its association with self-esteem was non-significant (β = 0.04; p > 0.05). Overall, effect sizes for HSPS tended to be higher than those for SPSQ in the fully adjusted models.

Table 3 presents the associations of SPS with the total GSES score, guilt and shame assessed separately, and self-esteem, adjusted for age, gender, education, and neuroticism. Individuals with medium SPS were approximately twice as likely to experience guilt and shame compared to those with low SPS. Respondents with high SPS demonstrated a greater likelihood of reporting high GSES total scores (odds ratio [OR] = 6.27; p < 0.001) and elevated guilt (OR = 6.38; p < 0.001), whereas the associations with shame and low self-esteem were non-significant. A similar pattern was observed when examining only sensory sensitivity (Model 2), in which participants with high sensory sensitivity were more likely to report heightened feelings of guilt (OR = 4.42; p < 0.001) and high GSES total score (OR = 4.54; p < 0.001).

Table 3.

Associations of sensory processing sensitivity with guilt, shame, and self-esteem: results of binary logistic regression (with the low SPS respondents being the reference category), crude, and adjusted for age, gender, education level, and neuroticism (odds ratios and 95% confidence intervals)

Model	Variable		SPS level	GSES total score	Guilt	Shame	Low self-esteema
1	SPSQ	Crude	Low	1	1	1	1
			Medium	2.79*** (1.53–5.07)	2.36*** (1.49–3.74)	1.59 (0.83–3.07)	1.49 (0.75–2.96)
			High	15.33*** (7.51–31.28)	14.33*** (7.45–27.55)	5.59*** (2.62–11.91)	4.76*** (2.14–10.6)
		Adjusted	Low	1	1	1	1
			Medium	1.99* (1.04–3.79)	1.64 (0.99–2.69)	1.2 (0.59–2.43)	0.79 (0.37–1.70)
			High	6.27*** (2.81–13.96)	6.38*** (3.08–13.25)	2.12 (0.9–5.02)	0.91 (0.36–2.35)
2	SPSQ-S	Crude	Low	1	1	1	1
			Medium	2.38*** (1.46–3.87)	2.16*** (1.45–3.2)	1.62 (0.91–2.87)	1.92* (1.00-3.7)
			High	8.76*** (4.53–16.91)	7.74*** (4.15–14.43)	4.55*** (2.18–9.51)	5*** (2.19–11.42)
		Adjusted	Low	1	1	1	1
			Medium	1.71 (1.00-2.94)	1.59* (1.03–2.47)	1.22 (0.66–2.27)	1.19 (0.57–2.47)
			High	4.54*** (2.13–9.69)	4.42*** (2.17–8.99)	2.13 (0.92–4.92)	1.55 (0.59–4.05)
3	HSPS	Crude	Low	1	1	1	1
			Medium	8.11*** (2.95–22.32)	3.55*** (2.02–6.24)	5.17** (1.61–16.62)	2.04 (0.87–4.81)
			High	44.18*** (15.24-128.07)	19.3*** (9.75–38.22)	21.38*** (6.39–71.51)	7.33*** (2.93–18.37)
		Adjusted	Low	1	1	1	1
			Medium	4.45** (1.57–12.64)	1.98* (1.09–3.61)	3.23 (0.97–10.69)	0.7 (0.28–1.78)
			High	14.84*** (4.83–45.58)	6.83*** (3.23–14.45)	8.02** (2.24–28.79)	1.03 (0.36–2.98)

SPSQ SPSQ total score, SPSQ-S Sensory Subscale score of the SPSQ, Guilt Guilt subscale of the GSES, Shame Shame subscale of the GSES

^aDichotomisation based on the reversed summary RSES score: a higher score indicates lower self-esteem

* p < 0.05, ** p < 0.01, *** p < 0.001

Using the HSPS to assess SPS (Model 3), the results indicated an increased likelihood of experiencing guilt and shame among individuals with high and medium SPS compared to low SPS participants, with ORs ranging from 1.98 (p < 0.05) to 14.84 (p < 0.001). For shame specifically, a significant association was found only in the high SPS group (OR = 8.02; p < 0.001). There were no significant associations with lower self-esteem.

Discussion

This study aimed to examine the relationships between SPS and negative emotional experiences, i.e., feelings of guilt, shame, and low self-esteem, while controlling for neuroticism. We found that SPS levels differed significantly by gender, education, and economic status. Overall, higher SPS values were significantly associated with increased feelings of guilt and shame, and low self-esteem. However, the association with low self-esteem became non-significant after adjusting for neuroticism, whereas the other associations remained significant. Similar patterns emerged when only sensory sensitivity was examined. Additionally, the use of an alternative SPS measure (HSPS) yielded comparable results, though the observed odds ratios were generally higher.

We observed that SPS levels differed significantly by gender, education, and economic status, with women being more prevalent in the high-SPS category relative to men. This finding aligns with the majority of large-scale studies reporting significantly higher total SPS scores among women [63]. Regarding educational attainment, we observed a significant overall association between education level and SPS category, although no post-hoc tests were performed to identify specific differences. The literature reports mixed findings regarding education and SPS [63]. Some studies report positive correlations between higher education and SPS [69, 71, 80], whereas others find negative or non-significant relationships [81]. Such discrepancies may reflect variations in sample characteristics, differences in measurement tools, or the complexity of the relationship between SPS and psychosocial factors that influence educational trajectories. We also found that SPS levels varied by economic status. On this topic, the evidence remains limited and inconclusive [69, 82]. These mixed findings may reflect both the limited number of studies on this topic and the nuanced ways in which economic conditions influence sensitivity. Limited resources may increase stress and sensitivity to threats, whereas greater resources may provide richer environments that heighten responsiveness [41]. Finally, we found no significant differences in SPS by age or marital status. These null findings are consistent with some earlier reports [71, 81], though further research is needed to explore how life stage, relationship roles, or aging processes may influence environmental sensitivity [63]. Overall, these findings suggest that sociodemographic characteristics may play a role in SPS expression, though the underlying mechanisms remain complex and warrant further investigation.

The proportion of HSPs in our sample (9.2%) was lower than estimates commonly reported in the literature, typically ranging between 15% and 30% [35, 48, 70]. This discrepancy may be related to several factors. First, prevalence rates may vary depending on the sample characteristics [83], including gender distribution. For example, women generally report higher SPS scores [35, 84], and thus, studies with a greater proportion of women often report a higher prevalence of high SPS. Second, the method used to define SPS categories can be substantially associated with differences in prevalence rates. Some studies have employed binary classification (low vs. high sensitivity) using fixed cut-off scores [e.g., 35, 82]. Recent work has adopted a three-level model (low, medium, and high sensitivity) using a latent-class approach [70, 85] or sample-based percentile cut-offs (e.g., top and bottom 30%) [86, 87]. Following the recommendation by Lionetti et al. [70], we divided our sample into tertiles according to total sensitivity score. We believe this method enhances transparency and reproducibility and makes results more comparable across studies.

Despite the use of different instruments to measure SPS, the relationship between SPS and both feelings of guilt and shame remained significant, with guilt showing a particularly strong link to SPS. Studies suggest that high levels of SPS may be associated with various mental health issues, such as anxiety or depression [88]. These conditions may increase an individual’s susceptibility to feelings of guilt and shame. Alternatively, enhanced information processing, as evidenced by increased EEG activity in HSPs [86], may lead to greater awareness and more intensive processing of negative emotional experiences. Another possible explanation for why sensitive individuals tend to experience more shame and guilt and possess lower self-esteem lies in early-life influences. Méndez Leal et al. [89] found that adversity in early care settings is associated with sensory processing difficulties, including excessive sensory reactivity, which may contribute to mental health problems mediated by perceived stress [90]. Research also suggests that HSPs may be more adversely affected by poor parenting, suggesting an interaction between SPS and environmental factors [55, 57], which is consistent with the theory of differential susceptibility [41].

Moreover, SPS is associated with empathy, as indicated by increased activity in the brain regions linked to self-awareness and empathic processing [38, 91]. Some theories propose that empathy may have played a pivotal role in the evolutionary emergence of guilt. For example, Decety and Holvoet [92] suggest that empathy may lead to feelings of guilt by enabling individuals to understand more fully the impact of their actions on others. Thus, empathy may serve as a bridge to understand the association between SPS and feelings of guilt.

The stronger association of SPS with guilt may be related to more intense reactions to both internal and external stimuli among HSPs [93]. SPS is linked to heightened sensitivity and emotional reactivity to environmental and social stimuli [38], particularly negative events experienced in daily life [37]. Therefore, SPS appears to involve greater sensitivity to specific actions rather than to one’s overall sense of self.

Importantly, we found that neuroticism played a distinct role in the associations between SPS and feelings of guilt and shame compared to SPS and self-esteem. The fact that the association with self-esteem disappeared once neuroticism was controlled for strongly suggests that neuroticism acts as a confounding variable in this specific relationship. Neuroticism, characterised by heightened negative emotionality, consistently demonstrates a strong inverse relationship with self-evaluations [94–98], and HSPs often exhibit elevated levels of neuroticism [51, 99]. Consequently, it is plausible that the variance in self-esteem initially attributed to SPS can be largely accounted for by this broader disposition for negative affect and self-criticism. In other words, it may not be the SPS itself that is related to self-esteem, but rather emotional instability (i.e., neuroticism), which often co-occurs with it. Importantly, SPS remains a theoretically distinct construct emphasising deep perceptual and cognitive processing [35], though in some of its aspects (as reflected by the Ease of Excitation and Low Sensory Threshold subscales of the HSPS scale) overlaps with neuroticism [100].

The finding that the associations between SPS and feelings of guilt and shame, although weakened, remained significant even after adjusting for neuroticism suggests a more complex relationship. This attenuation indicates that neuroticism may function as a substantial confounding variable, or potentially as a mediator. Therefore, these associations cannot be explained solely by emotional instability, but rather by essential aspects of SPS that are independent of neuroticism, such as heightened interpersonal awareness, empathic responsiveness, or moral sensitivity [38, 92]. For example, HSPs’ heightened perception of others’ emotions can intensify their feelings of guilt and shame by increasing their sensitivity to causing harm or failing to meet expectations of others. Furthermore, deeper cognitive processing [35] may lead HSPs to think more intensely about their actions and their consequences, thereby increasing feelings of guilt. Overall, our findings highlight the necessity of controlling for neuroticism when examining SPS correlates and raise the possibility that neuroticism partly explains, but does not fully account for, the link between SPS and guilt/shame. This hypothesis, however, requires formal testing through mediation analysis in future longitudinal research.

Analysis based on different levels of SPS consistently revealed a similar pattern of results, with the strongest associations among respondents in the high-SPS category. Moreover, results differed slightly depending on the instruments used (SPSQ, SPSQ-S, and HSPS). Specifically, the association with shame was found only in the fully adjusted model for the HSPS, but not for the SPSQ. These discrepancies may be explained by the broader scope of the HSPS, which extends beyond sensory sensitivity to include interrelated factors such as conscientiousness, being easily surprised and having a rich inner life [83].

Conversely, significant associations were also found when examining only the sensory aspects of SPS (as measured by the SPSQ sensory subscale). This may indicate that the link between SPS and negative emotions is not solely attributable to broader personality traits. For instance, heightened sensory sensitivity itself may be a relevant factor. Specifically, individuals with heightened auditory sensitivity may perceive and interpret verbal cues and communication with greater intensity. When confronted with criticism or negative verbal expressions, they may process such information more intensely and respond more emotionally, including experiencing guilt related to perceived mistakes.

Furthermore, dividing respondents into three groups based on their SPS levels revealed interesting dynamics in the relationship between SPS and negative emotional experiences. Although a linear trend might have been anticipated, our data suggest that individuals in the highest-SPS group reported considerably higher scores than those with lower SPS levels. These findings highlight the potential value of categorising respondents based on their SPS levels. This methodological consideration suggests that categorising SPS scores may provide valuable insights not only into emotional reactivity and negative emotional experiences but also across other life domains.

Implications

The study’s findings have important implications for understanding the relationship between SPS, emotional states (guilt, shame, and self-esteem), and the influence of neuroticism. Clinicians can use this knowledge to tailor interventions for individuals with high SPS, helping them manage heightened emotional responses more effectively. The apparent explanatory role of neuroticism in the link between SPS and self-esteem underscores the importance of addressing neurotic tendencies within self-esteem interventions. Public health initiatives should take into account the mental health implications of SPS. Future research should further explore the complex interplay between SPS, guilt, shame, self-esteem, neuroticism, early life experiences, and emotional well-being. In educational and community settings, greater awareness of SPS may foster more inclusive and supportive environments for sensitive individuals, contributing to their overall well-being.

Strengths and limitations

This study has several notable strengths. First, it examines the direct associations between SPS, feelings of guilt and shame, and self-esteem, while accounting for the influence of neuroticism. Second, the sample closely reflects the demographic structure of the national population in terms of age and gender. Third, the use of two independent instruments to assess SPS, including one validated in the Czech context, strengthens the robustness of measurement and supports cross-method reliability. However, several limitations should be acknowledged. Most importantly, the cross-sectional design of the study precludes any conclusions about causality. Although our findings suggest that neuroticism may account for the association between SPS and self-esteem, this cannot be directly tested within the limits of cross-sectional data. To establish temporal sequencing and causal relationships, future research should employ longitudinal or experimental designs. Another limitation arises from potential information bias due to reliance on self-reported data, which may be susceptible to social desirability bias. Furthermore, the online nature of the questionnaire introduces potential sampling bias by excluding individuals without internet access.

Conclusions

We found that higher SPS scores were significantly associated with increased feelings of guilt and shame and lower self-esteem, even after adjusting for demographic variables. However, when neuroticism was included in the models, the associations with feelings of guilt and shame remained significant, whereas the link with self-esteem was no longer statistically significant. Group comparisons based on SPS levels (using the SPSQ) revealed that individuals with high SPS were more likely to experience guilt and a combination of guilt and shame, but not shame or reduced self-esteem alone. Similar associations emerged using the HSPS, showing even stronger links between higher SPS and feelings of guilt and shame.

Abbreviations

SPS

Sensory Processing Sensitivity

HSP

Highly Sensitive Person

SPSQ

Sensory Processing Sensitivity Questionnaire

SPSQ-S

Sensory Sensitivity Subscale of SPSQ

HSPS

Highly Sensitive Person Scale

GSES

Guilt and Shame Experience Scale

GSES-G

Guilt Subscale of GSES

GSES-S

Shame Subscale of GSES

RSES

Rosenberg Self-Esteem Scale

BFI-N

Big Five Inventory – Neuroticism Subscale

OR

Odds Ratio

CI

Confidence Interval

Authors’ contributions

MB and KM conceptualized the study; MB wrote original draft preparation; MB, KM, MCB, VH, and PT reviewed and edited the paper; MB conducted statistical analyses under the supervision of KM and PT; MB, KM and MCB interpreted results; VH and PT supervised the study; KM, VH and PT secured funding and resources. All authors contributed to manuscript revision, read, and approved the submitted version.

Funding

The work was supported by ERDF/ESF project DigiWELL (No. CZ.02.01.01/00/22_008/0004583).

Data availability

The datasets generated and/or analysed during the current study, the study code, as well as supplementary materials, are available in the Open Science Framework (OSF) repository, https://osf.io/nc7dr/.

Declarations

Ethics approval and consent to participate

The study design received approval from the Ethics Committee of the Faculty of Theology at Palacký University in Olomouc (No. 2020/06). Informed consent was obtained from all participants. All research was performed in accordance with the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.