How Sensory Processing Sensitivity Shapes Employee Reactions to Core Job Characteristics

Sensory Processing Sensitivity: Definition and Theoretical Framework

In their seminal work, Aron and Aron (1997) introduced the concept of SPS and proposed that SPS consists of four facets of sensitivity: (1) depth of information processing, (2) ease of overstimulation, (3) increased empathy, and (4) greater awareness of environmental subtleties. Aron and Aron (1997) also developed a 27-item measure to assess SPS, the Highly Sensitive Person Scale. More recent findings by Smolewska et al. (2006) suggest that SPS, as measured by the Highly Sensitive Person Scale, comprises three facets of SPS instead of the four dimensions initially described by Aron and Aron (1997). The three facets are low sensory threshold (LST; i. e., being easily overwhelmed by things like bright lights, strong smells, or noisy sounds), ease of excitation (EOE; i. e., being mentally overwhelmed by external and internal stimuli like the mood of other people), and aesthetic sensitivity (AES; i. e., awareness of aesthetic like being deeply moved by arts or music). Generally, people who score higher in SPS tend to have a higher emotional reactivity and behavioral inhibition and higher awareness to “pause and check” in novel situations (Aron & Aron, 1997). A pause to check system means initially withdrawing from acting in each situation to assess it before a reaction is initiated (e. g., a decision is made). While a large body of research on SPS focused on its negative aspects, showing its correlations with negative psychological states, anxiety, depression, or burnout (Evers et al., 2008; Liss et al., 2008), fewer studies investigated more favorable positive consequences of SPS such as empathy and emotional intelligence (Sülzenbrück & Töpfer, 2019).

The construct of SPS is also a part of the environmental sensitivity concept by Pluess and Boniwell (2015), which also includes differential susceptibility (Belsky & Pluess, 2009) and biological sensitivity to context (Ellis & Boyce, 2011). These theories generally describe interindividual differences in the responsiveness to positive and hostile environments. The concept of differential susceptibility comprises the diathesis–stress model and the vantage sensitivity model. On the one hand, the diathesis–stress model proposes that higher sensitivity leads to higher vulnerability when encountering negative environmental stimuli. On the other hand, the vantage sensitivity model assumes that positive stimuli have a more pronounced positive effect on people higher in SPS (Pluess & Boniwell, 2015). The biological sensitivity to context model suggests that children respond differently to positive or negative stimuli depending on psychobiologic reactivity to stress. Thus, more reactive children show higher sensitivity to the environment than others (Ellis & Boyce, 2011).

In the present study, we transferred the assumption of the differential susceptibility concept (Belsky & Pluess, 2009) and SPS to the organizational setting. Instead of classifying participants as highly sensitive persons (HSPs) or non-HSPs, we considered SPS a continuous variable with interindividual differences from lower to higher SPS values on the measurement scale. We assumed that employees higher in SPS benefit even more from job characteristics generally seen as job resources than do employees with lower SPS.

Organizational View of SPS: Organizational Citizenship Behavior and Work Design

Knowing the potential benefits of having employees or leaders scoring higher in SPS is interesting for organizations, for example, concerning the employees’ attitude and behavior. Nevertheless, SPS has gained only little attention in organizational research in the last two decades. A few studies focused on the diathesis–stress approach and, therefore, on the negative organizational outcomes of SPS. In this line of research, Evers et al. (2008) found a positive association between SPS and burnout symptoms. However, work stress was only associated with two of the SPS subscales: LST and EOE. There was no such effect for the third facet, AES that does not have any content-relatedness to burnout symptoms. Redfearn et al. (2020) showed that SPS predicts stress and burnout among nurses. Andresen et al. (2018), furthermore, showed that the perceived stress fully mediated the positive effect of SPS on turnover intentions in a sample of expatriates.

However, empirical findings about the potential positive effects of SPS in organizations are rather sparse (e. g., Sülzenbrück & Töpfer, 2019) and did not consider its impact on OCB) as a positive organizational outcome so far. Organ (1988) defined OCB as “individual behavior that is discretionary, not directly or explicitly recognized by the formal reward system, and that, in the aggregate, promotes the effective functioning of the organization” (Organ, 1988, p. 4). OCB comprises different aspects of extra-role behavior such as helping behavior, sportsmanship, organizational loyalty, organizational compliance, individual initiative, civic virtue, and self-development and is associated with increased individual and organizational performance as well as organizational success (Podsakoff et al., 2000). In the context of research on substitutes for leadership, Podsakoff and MacKenzie (1995) identified job characteristics such as feedback as one of the antecedents of OCB.

The job characteristics model proposed by Hackman and Oldham (1976) provides the theoretical background for effects of task characteristics on work-related attitudes and organizational outcomes (e. g., OCB). The model proposes that five core job characteristics, namely, autonomy, skill variety, task significance, task identity, and feedback, are antecedents to the experience of three critical psychological states at work: experienced meaningfulness, experienced responsibility for work outcomes, as well as knowledge of the results of work activities. These psychological states, in turn, have positive effects on job attitudes such as job satisfaction and various job-performance measures, including OCB. Various studies provided empirical evidence for the assumptions of the job characteristics model in general (see, e. g., Fried & Ferris, 1987). More specifically, positive correlations between job characteristics and OCB were also found (Chiu & Chen, 2005; Nielsen et al., 2009).

Our first attempt to elucidate the patterns of effects of SPS on employee attitudes and behavior focused on aspects of work design. According to differential susceptibility (Belsky & Plues, 2009), employees higher in SPS might benefit even more from positive aspects of job design, leadership, and/or organizational culture compared with those lower in SPS. The former could therefore react to such positive work settings with even higher job satisfaction, commitment, and productivity. Our study extends previous research on SPS within organizations by investigating its moderating effect on the relationship between core job characteristics (Hackman & Oldham, 1976) and OCB.

Hypotheses

On the basis of job characteristics model (Hackman & Oldham,1976), postulating that job characteristics lead to positive work behavior like OCB, and studies confirming these assumptions (Chiu & Chen, 2005; Nielsen et al., 2009), we expected to replicate the positive effects of job characteristics on OCB. The vantage sensitivity model (Pluess & Boniwell, 2015) assumes that positive environmental stimuli (e. g., positive job characteristics) have a more pronounced positive effect on people higher in SPS. Therefore, we propose a strengthening influence of a higher SPS on associations between job characteristics and OCB. Here, we focused on the five job characteristics proposed by Hackman and Oldham (1976) in their job characteristics model: autonomy, task identity, task significance, task variety, and feedback.

Participants, Design, and Procedure

Our sample (N = 199 participants; 42 male, 157 female) was a convenience sample and consisted of 174 full- or part-time employees in various types of industries and 25 students. All participants were part-time students at a German university of applied sciences (evening or weekend courses) with 33 university centers located throughout Germany. They received course credit for taking part in the online survey. The mean age of participants was M = 27 years (SD = 4.39). The sample comprised mostly higher-educated non-management employees from the service industry and other industries (see Table 1 for biographical information on the participants).

Table 1 Biographical information of participants

Index	N
Gender
Male	42
Female	157
Educational attainment
Higher education entrance qualification	120
Subject-related entrance qualification	58
Intermediate school-leaving certificate	18
University degree	3
Career status
Employee	171
Student	25
Official	1
Trainee	1
Freelancer	1
Sector
Other	55
Service industry	54
Finance and insurance industry	34
Automotive industry	17
Health and social services	11
Public administration	10
Construction industry	7
Energy sector	5
Hotel and gastronomy	4
Education and teaching	2
Management position
No	176
Yes	23

Note. Most selected categories are highlighted in bold. N = 199.

Table 1 Biographical information of participants

View as image HTML

To address common-method bias, we carried out a two-wave study, as proposed by Podsakoff and colleagues (2003). Participants’ self-reports of all predictor variables were assessed in the first wave (N = 315). The dependent variables and the moderator variable were measured in the second wave (N = 207), which was conducted at least 6 days after the first wave. The drop-out rate between the first and second wave was 34.29 %. The data from eight participants had to be excluded (see “Data Analysis”), so that data from 199 participants were included in the analysis.

Measures

SPS was measured with the German translation of the HSP scale (HSP-G) of Aron and Aron (1997), developed and modified by Konrad and Herzberg (2019). The scale consists of 26 items (e. g., “Stimmungen anderer Menschen beeinflussen mich” [“Other people’s moods influence me]), comprising three subscales proposed by Smolewska et al. (2006): SPS_EOE (α = .81), SPS_LST (α = .78), and SPS_AES (α = .72). A 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree) was used.

To measure job characteristics, we used the respective subscales for Task Characteristics (work-scheduling autonomy, decision-making autonomy, work methods autonomy, task variety, task significance, task identity, and feedback from job) of the Work Design Questionnaire (WDQ; Morgeson & Humphrey, 2006) in the German version developed by Stegmann et al. (2010), for example, “Ich kann meine Arbeit so planen, wie ich es möchte” (“I can plan my work the way I want to”; Item 3 of the subscale work-scheduling autonomy). A 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree) was used. Cronbach’s α values for the subscales were: work-scheduling autonomy (α = .80), decision-making autonomy (α = .90), work methods autonomy (α = .83), task variety (α = .78), task significance (α = .79), task identity (α = .70), and feedback from job (α = .62; Stegmann et al., 2010).

To assess OCB, the OCB scale by Podsakoff et al. (1990) was used. We used the back‐translation method by Brislin (1970) to translate into German. The scale consists of 24 items (e. g., “Ich mache keine zusätzlichen Pausen” [“I do not take extra breaks”]) and was assessed with a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). Cronbach’s α .80 (Podsakoff et al., 1990).

Control variables were gender and age, which were included in the first step of the multiple hierarchical regression analyses. In this study, age was used as a metric, and gender as a nominal variable. Since previous studies (e. g., Konrad & Herzberg, 2019) have shown potential differences in SPS between men and women as well as differences regarding their age, we included these variables to control for potential age and gender effects.

Data Analysis

For the statistical analyses, we used SPSS Statistics Version 26. Only participants completing both survey parts were included in further analyses (N = 207). Before conducting the statistical analyses, the data quality was checked, excluding potentially meaningless, careless, or fraudulent responses via the index score Time-RSI proposed by Leiner (2019). Participants with index values greater than 2 and datasets with missing values were excluded (N = 8). Finally, data from 199 participants were included in the statistical analyses.

We computed multiple hierarchical regression analyses in three steps to test our hypotheses. The first step included the control variables gender and age; in the second step, the main effect of the respective work design facet was included; and finally, in Step 3, we tested the main effect of SPS and the moderator effects. All analyses were conducted for the overall mean score of SPS as a moderator as well as for its subscales SPS_EOE, SPS_LST, and SPS_AES separately. Figures depicting the moderating effects were created using the templates provided by Dawson (2021).

Autonomy

The exploratory analyses of autonomy with the SPS subscales revealed a significant moderating effect for the relationship between autonomy and OCB for the subscale SPS_LST, B = −.11, p = .033, f² = .28, F‍(193) = 3.70, R² = .09 (see Figure 3). By contrast, there were no significant moderating effects for autonomy in total with the subscales EOE and AES. The exploratory analyses of the autonomy subscales with the SPS subscales revealed significant moderating effects for the relationship between work methods autonomy and OCB for the subscales SPS_EOE, B = −.13, p = .033, f² = .28, F‍(192) = 4.06, R² = .07 (see Figure 5) and SPS_LST, B = −.14, p = .008, f² = .29, F‍(193) = 4.34, R² = .08 (see Figure 6). There were no further significant moderating effects. In line with the analyses of autonomy (H1) reported above, the effects for the subscales of SPS were also all negative (see Figure 4 – 6).

Task Variety

For task variety and OCB, a negative moderating effect was found for the subscale SPS_EOE, B = −.10, p = .048, f² = .49, F‍(191) = 10.36, R² = .19 (see Figure 7). There were no further significant moderating effects.

Task Identity

For task identity and OCB, we did not find any significant moderating effects.

Task Significance

For task significance and OCB, the analysis showed a positive moderating effect for the SPS subscale AES, B = .20, p = .001, f² = .31, F‍(192) = 4.92, R² = .09 (see Figure 8). There were no further significant moderating effects.

Feedback From Job

For the relationship between feedback from job and OCB, we found another significant positive moderating effect of the SPS subscale AES, B = .11, p = .049, f² = .18, F‍(192) = 2.23, R² = .03 (see Figure 9). There were no further significant moderating effects. In line with the analysis of feedback from job (H5) reported above, the relation between feedback and OCB was strengthened by SPS, especially for the facet SPS_AES.

In summary, SPS strengthened the relationships between task significance and OCB (H3) as well as feedback from job and OCB (H5; see Figures 2, 8 and 9), whereas SPS weakened the effect of autonomy on OCB (H1; see Figures 1 and 3 – 6).

Practical and Theoretical Implications

Our results indicate that SPS influences the relationships between job design characteristics and positive work behavior such as OCB. Contrary to our assumptions, these associations are not all positive, but altogether are valuable to be considered by organizations. Our findings highlight the need for companies to ensure that their working conditions are adapted to their employees’ individual needs and skills. For example, as the results of our first hypothesis show, the higher sensitive that an employee is, the more pronounced the negative effect on the relationship between autonomy and OCB. Under working conditions with a higher degree of autonomy, especially in work methods, employees with higher SPS produce less additional engagement of their own volition.

However, our results also show that especially these employees with higher SPS benefit from resourceful working conditions such as the significance of their task or feedback from their job. Under these favorable working conditions, they show even more extra-role behavior than less sensitive individuals. However, the notion that “good” working conditions concerning work design in a “one size fits all” approach is challenged by our findings showing that for more sensitive people, typical beneficial working conditions such as task variety and especially autonomy can harm their work attitudes and behavior. Therefore, managers need to be able to assess their employee’s sensitivity levels and utilize this knowledge to adjust working conditions individually. Consequently, SPS might also be another relevant factor added to organizations’ broad spectrum of diversity aspects.

Addressing the theoretical implications of our findings, our results imply that SPS should be acknowledged as a personality trait moderating the relationship between task characteristics and organizational outcomes. The notion that some aspects of work design, such as autonomy and task variety, are often interpreted as job resources is challenged by our findings, indicating the need for an individualized assessment of work conditions before categorizing them as job resources or job demands. Our findings on SPS, therefore, extend previous research on personality traits moderating the associations between job design characteristics and organizational outcomes (e. g., De Jong et al., 2001).

Limitations and Future Directions

There are several limitations to our study. For example, although adequate for regression analyses, the sample size (N = 199) might have not been sufficient to detect smaller effects. Additionally, the cross-sectional design might have resulted in a common-method bias (Podsakoff et al., 2003). We addressed this concern by first introducing a time lag between the assessment of our variables. Second, we conducted the Harman one-factor test (Harman, 1976). For this test, the principal component analysis of all variables is computed. A common-method bias has likely affected the data if the first extracted factor explains more than 50 % of the observed variance. We found that the first extracted factor explained only 12 % of the observed variance, being far below the critical value of 50 % of explained variance. This result indicates that the common-method bias did not affect our results. Limitations also occur from assessing all variables as self-reports. Although it has been shown that self-reported performance measures are comparable to other sources (Facteau & Craig, 2001), further studies replicating our findings should include multisource ratings.

Although we were able to recruit employees from a broad range of industries, there are some limitations due to the convenience sample. Since all participants were quite young (mean age of M = 27 years, SD = 4.39) and mostly female (79 % of the participants) part-time or full-time students earning their bachelor’s degree at a university of applied sciences, the generalizability and external validity of our results is constrained and further research investigating the impact of SPS in other populations is needed. However, we aimed to account for the potential effects of age and gender by controlling for these two demographic variables in the hierarchical regression analyses.

Since the data were collected in an online-based survey study, objectivity can be assumed for our study. Concerning reliability, most of the measures we applied showed at least acceptable values for Cronbach’s α, ranging from α = .77 for Feedback from Job to α = .90 for SPS. However, one of the subscales of SPS (AES) only reached α = .60 in our data analyses. Therefore, the results, especially concerning this facet, should be interpreted with caution and replicated in further research.

Although we found several significant moderating effects of SPS and its facets, the size of the observed effects was relatively small, with the increase in explained variance due to SPS and its facets ranging between ∆R² = .02 (AES and feedback from job) and ∆R² = .07 (LST and work methods autonomy).

Moreover, our results provide further evidence that it is crucial to investigate SPS both holistically and in its facets separately, considering the somewhat contradicting mechanisms underlying its three facets.

Considering future directions, further attempts to optimize the SPS measure should be undertaken, particularly addressing its dimensionality and the negative wording of various items. According to Benham (2006), Ahadi and Basharpoor (2010), and Bakker and Moulding (2012), SPS seems to be a personality trait with an overall negative impact on various outcome variables, but not for all its subscales. Most of these studies suggest that SPS has two main facets, a negative facet comprising LST and EOE as well as a positive facet (AES). Wyrsch (2020) called these two facets vulnerable sensitivity (EOE and LST) and vantage sensitivity (AES), finding that vulnerable sensitivity was associated with decreased OCB, while vantage sensitivity increased OCB. As our results indicate, AES may be a resourceful aspect of this trait and, therefore, might represent vantage sensitivity (Wyrsch, 2020), while EOE and LST reflect vulnerable sensitivity. In line with Wyrsch (2020), we found negative moderating effects for the two construct facets of SPS relating to vulnerable sensitivity (EOE and LST) as well as for the overall score of SPS on the relationship between autonomy and task variety with OCB. Additionally, our results show significant positive moderating effects for vantage sensitivity (AES) on the relationships between task significance and feedback with OCB. Future research should focus on the subscales of SPS, notably the AES subscale, and revisit the construct clarification of SPS, potentially adding more information about the different facets of higher sensitivity via qualitative research approaches.

The investigation of core job characteristics is only one of many approaches to studying the role of SPS in the impact of work design. From a broader perspective, the concept of job crafting, “…changing cognitive, task and/or relational boundary to shape interactions and relationships with others at work…” (Wrzesniewski & Dutton, 2001, p. 179), might be a promising research path worth following in order to understand the role of SPS in creating a work setting that fits the needs of people higher in SPS in a more proactive way.

Our results revealed that employees with higher SPS benefit from certain working conditions to achieve their best performance and thrive in the workplace. Aside from work design and job characteristics, other organizational attributes could influence people higher in SPS, for example, the type of experienced leadership behavior. In a first attempt to connect SPS with leadership, Panetta (2017) examined subjective leadership theories of highly sensitive leaders within organizations. Sülzenbrück and Töpfer (2019) provide a theoretical framework proposing that, due to their increased empathy and emotional intelligence, people scoring higher in SPS could be more likely to emerge as leaders and to be successful leaders as well. To access these potentials lying in higher sensitive employees and/or leaders, a work design that addresses their individual needs is required.

Conclusion

Our findings emphasize that SPS has an impact on organizational factors such as OCB. On the one hand, SPS strengthened the positive impact of task significance and feedback from the job on OCB. On the other hand, adverse moderating effects of SPS were found for the relationship between autonomy (especially in work methods) and OCB, as well as for task variety and OCB. Furthermore, the pattern of results found for the subscales of SPS indicates that the AES subscale could be considered a resource facet of this personality trait in terms of vantage sensitivity (Wyrsch, 2020), while EOE and LST reflect vulnerable sensitivity. Our findings shed more light on the understanding of SPS and inspire even more research on SPS in the organizational context.