Can High Levels of Hindrance Demands Increase the Worker’s Intellectual Response?

Job demands are factors that are associated with a physical and psychological cost when it comes to coping with them, but which can also positively affect the motivational process. Demands such as overload, defined as an excessive workload, have not presented positive results in any of the studies that have related it to employee engagement. The present study aims to delve into the possible positive effect of this demand on the intellectual bonding of employees. It is hypothesized that: (a) Initially, the increase in the perception of work overload will show a negative influence on the intellectual engagement of the employees; but (b) high perceptions of overload will change this effect, producing an increase in the intellectual dimension of engagement. The sample is made up of 706 employees of a Spanish multinational company. The results support this asymmetric curvilinear influence. The level of intellectual engagement is significantly reduced when the role overload increases from the lower values of the scale. However, upon reaching high levels of role overload, the intellectual engagement response begins to grow. These results challenge the conceptualization of overload as only negative and opens the door to consider that the positive response to a demand can also occur at high levels of it.


Introduction
Globalization and technological progress have generated great changes in the context of work, assuming an increase in job demands [1]. Various studies have coincided in pointing out the work overload as the demand with the greatest negative consequences at the economic level, its effects being equivalent to the loss of 3.5% of the Gross Domestic Product (GDP) of a country [2]. In this context, it could be said that the perception of being overloaded has become one of the most important work limitations today [3,4].
Role overload occurs when excessive demands on time and energy exceed the individual's available time and energy [5]. Therefore, two forms of role overload can be distinguished: First, when an employee's workload is too heavy to complete on time, and another occurs when employees lack the necessary skills to complete a task well [6].
Work overload not only has economic consequences, but it also affects the employees themselves. Research has repeatedly associated the perception of role overload with negative effects for employees, such as sleep disturbances, the development of emotional disorders, and the deterioration of physical health [7]. Therefore, this job demand has also been identified as one of the most important impediments to professional development and well-being in employees [8], with a special incidence in the employees of multinational companies [9]. Researchers interested in delving into international and multicultural organizations must understand the challenges these workers face in the workplace in order to develop interventions that improve the well-being of these workers. [10].
In recent years, the interest in deepening the study of job demands has experienced an important boom [11,12]. This has been possible thanks to the development of theoretical models and research that have proposed that labor demands do not have to have exclusively negative consequences [13]. Authors such as Demerouti and Bakker [14] or Podsakoff, LePine and LePine [15] propose an approach to the study of demands, understanding them also as a challenge, so that they can generate positive results both in professionals and in the organization.
One of the key elements in relation to role overload is employee engagement. The time and energy devoted to a role can reflect the individual's engagement with that role, as well as their general availability of time and energy [16]. According to Marks and MacDermid [17], role balancing is both a behavioral pattern of acting through roles in a certain way and a cognitive-affective pattern. From this perspective, work environments with a certain level of demands could lead to an increase in positive motivational states and job performance [18]. Therefore, the positive balance of roles can affect the tendency to become intellectually involved in the performance of each role in one's total role system, to approach each typical role and role partner with an attitude of attention and care [17]. However, in the study of job demands, there are still many unknowns to solve: Can all demands be challenging? If so, how do hindrance demands become challenging demands or vice versa? and, what factors can affect the transition of a demand between obstacle and challenge?
The present study aims to deepen the relationship between the hindrance demands and the motivational process, specifically studying the effect of high and low values of overload on intellectual engagement. To do this, it tries to test the hypothesis of whether role overload can have a positive influence on the emotional response of employees. It is expected that, in work contexts with increasing levels of role overload, employees initially experience a reduction in their level of intellectual engagement. However, when faced with high overload values, this influence could change, generating a positive effect that increases intellectual engagement. This supposes a novel vision of the influence of the demands, especially an asymmetrical curvilinear relationship that captures the complexity of job demands contexts.

Job Demands: Hindrances or Challenges
When studying the influence of the organizational context on employee responses, there are two concepts that have focused the attention of researchers in recent years: job demands and resources [19]. From the Job Demands-Resources Theory (JD-R), it is proposed that both the demands and the resources are the starting point of the processes of deterioration of health and motivation, respectively, although the influence between both processes is constant [18].
Focusing on job demands, these are defined as factors of the work context whose coping involves a cost, and which reduce the ability of people and organizations to achieve their goals [14,20]. These were initially conceived as elements with negative consequences [21]. However, in a recent review, Bakker and Demerouti [13] proposed the possibility that demands can become challenging factors with positive consequences, promoting growth and personal achievement. These challenges would be perceived as rewarding experiences that are well worth the discomfort generated, which is why they are considered "good" stressors [20,21].
Current research proposes that the same demand can act as an obstacle or as a challenge depending on contextual factors such as leadership [21] or the personal resources available to the employee [22,23]. Although a key element in considering a demand as an obstacle or a challenge seems to be the level at which the demand itself is perceived by the worker [24].
Among job demands, one of the most relevant and studied is role overload [3,4]. The perception of a role overload has shown negative effects on employees such as increased sleep disorders, reduced cognitive abilities, and deterioration in physical health in general [7]. At this point, it is necessary to clarify the difference between workload and role overload. The workload refers to the number of tasks that have to be attended in a job, which can increase or decrease, while role overload identifies the employee's perception of an excess in workload levels. This distinction is important, since when the workload is considered excessive, it will have negative consequences from its lowest values, so it must be considered a clearly hindering variable [25].

Connecting Demands to Intellectual Engagement
The psychological bond of the employee with the organization has been shown to only occur when employees feel that the organization favors and values its resources [26]. A relevant approach to the study of engagement is that carried out by Soane et al. [27] based on the satisfaction of needs. Soane proposes a measurement of this variable through the assessment of three dimensions: intellectual, affective, and social. The first dimension refers to the level of intellectual absorption at work and the contribution of ways to improve it; the second dimension refers to the state of positive affect that is experienced related to the job role; and the third dimension introduces the level at which the employee connects socially with the work environment and shares common values with colleagues.
This study focuses on the dimension of intellectual engagement, due to its special link with overload. Authors such as Khalid et al. [28] conceptualize this dimension as the degree of effort made by employees to contribute their knowledge and intellectual capacity when performing the work tasks associated with a role more effectively and efficiently. Research has shown its relevance as it is the first of the three dimensions of engagement to increase its values during the employee's emotional bonding process, which later extends to affective and social engagement [29].

Asymmetric Curvilinear Hypotheses
The interest in maximizing the health and well-being of employees has been observed in previous empirical studies [30], but most of them proved a linear relationship between variables (for example, proposing positive links from labor resources to positive emotions of the worker). As a general rule, research results have shown that employees who perceive their resources as insufficient when responding to demands tend to reduce their levels of engagement [31]. This leads us to think that less overloaded work contexts will be more likely to strengthen and increase employee engagement. However, these linear relationships may not capture the complexity of human behavior in an organizational context. Human beings are sensitive to situations of imbalance because they violate an important rule of the moral code in our societies.
Based on the Yerkes-Dodson theory [32], Quick et al. [24] show other hypotheses alternatives to explain the relationship between job demands and intellectual engagement. For example, the equity hypothesis [33] states that only a balanced perception of job demands is capable of producing the highest levels of well-being and supposes a symmetric curvilinear relationship between variables. However, the nonlinear relationship may present another form that would fit into an asymmetric hypothesis, due to the special sensitivity of humans to negative events [34], which in the present study are associated with perceptions of high role overload. In this sense, the initial increase in demands would also mean an increase in their level of engagement, thus acting as a challenging demand. However, at a certain moment, the excitement and activation derived from the progressive increase in demand will generate exhaustion, and engagement begins to decrease [35], which turns it into a hindrance demand.
Role overload will cause a decrease in employee engagement from the first moment [35]. Thus, in line with the Yerkes-Dodson theory, a context with role overload will be located from the first moment in the decreasing zone of the hood, causing a reduction in the effort of intellectual absorption and interest of employees in contributing their knowledge as employees in an attempt to minimize the net loss of resources [36]. This leads to the first of the hypotheses.
Hypothesis 1 (H1). Workers subjected to increasing role overload environments will initially experience a reduction in their level of intellectual engagement with the company.
However, faced with a context in which role overload is increasing, is the only possible answer a reduction in intellectual engagement, or can that trend change at some point? There are few studies that propose an increase in engagement in a context where demands reach high levels, and most of them do not refer to organizational contexts. For example, Billings, Folkman, Acree, and Moskowitz [37] studied the response of caregivers of people with AIDS to high emotional demands, finding that they maintain their positive emotional states, and even increase their level at the times of greatest demand. These authors propose that this response is intended to be a self-protective reaction, the purpose of which is to slow down the process of emotional deterioration so as not to be affected by the physical and emotional symptoms associated with it. In a similar way, and applied to the organizational context, Seers, McGee, Serey, and Graen [38] had already shown the use of organizational resources as a self-protective element. In this study, they conclude that employees, when faced with higher levels of role conflict, showed a greater need for supportive behaviors. More recently, Bakker et al. [39] follow this line and leave open the possibility that, under highly stressful conditions, people can use coping responses to reduce, not so much the stress of the organization, but the personal consequences that arise from it. These results, together with the Yerkes-Dodson theory (32), suggest the possible appearance of a possible increase in the worker's intellectual engagement in the face of higher values of role overload as a protective element, for which the following hypothesis is proposed.

Hypothesis 2 (H2).
The relationship between role overload and intellectual engagement will not be linear, but curvilinear, generating a change in trend that will increase the perception of the latter in employees with a higher perception of overload.

Summary of Novelty
This study aims to enrich the understanding of the role of job demands in the organizational context. The effect of labor demands is changing in the labor context. These can show more than one role depending on the level at which they are perceived. At the beginning, it will act as a challenge with positive results; by increasing its level it will become an obstacle to which the worker will respond negatively, but if it continues to increase its level of perception, it will give positive responses, although of less intensity.

Procedure
The sample is made up of 706 employees (response rate: 82.3%) of a multinational company dedicated to the production and distribution of materials for architecture and design, whose headquarters are located in the province of Almeria (Spain). All participants gave their informed consent before participating in the study. The study was carried out in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of the University of Almeria on 26 October 2016. Once they agreed to participate in the project, the employees of each department were informed by their managers about the objective of the study and its relevance to the organization. The questionnaires were administered in group sessions during the working day and at the company's facilities. The confidentiality and anonymity of the participants were ensured, since the use of alphanumeric codes replaced the identifying characteristics in the questionnaires.

Sample
The sample distribution included 22.6% of individuals between 18 and 25 years old, 31.2% between 26 and 35 years old, 29.8% between 36 and 45 years old, 15% between 46 and 55 years old, and 1.4% older than 56 years. Regarding gender, 91.2% of the sample were men, and 59.1% had a baccalaureate degree, 22.4% had a bachelor's degree or equivalent, 8.3% had university training, 3.4% had postgraduate or doctoral training, and 6.8% said they had other studies. In terms of tenure, 64.4% of the participants had less than five years in the company, 12.2% had between five and 10 years, 13.9% had between 10 and 15 years, and 9.2% had over 15 years. The most represented contract was undefined for 62.5% of the participants, while 35.4% had a temporary contract, and 2.1% had other types of contracts.
The only selection criterion for participants was that they should have a job in the company in which the questionnaire was applied at the time of the study.

Role Overload
To measure role overload, the dimension of the same name was used in the tool developed by Rizzo et al. [40], adapted and validated for use in Spain by Peiró et al. [41]. This scale consists of three items and asks whether workers can manage their current overload, the current pace of work required for the job, and to what extent they had to exert more effort than before to get the job done ("The amount of work that I have to do (or what they ask me to do) is excessive"). The response options were presented on a Likert scale with five categories, ranging from 1 (strongly disagree) to 5 (strongly agree), following the original design of the scale, with higher scores indicating a higher level of overload. Cronbach's Alpha for this scale was 0.86.

Intellectual Engagement
To measure this dimension, the Spanish version of the ISA Engagement Scale [42] was used. This scale is based on the measure developed by Soane et al. [27] that consists of three elements that ask the degree of focus within the workplace (for example, "I pay a lot of attention to my work"). The response options were presented on a Likert scale with seven categories, ranging from 1 (strongly disagree) to 7 (strongly agree), following the original design of the scale, with higher scores indicating a higher level of commitment. Cronbach's alpha for this scale was 0.94.

Control Variables
Because the evaluation of the work context is sensitive to the sex and age of the employees [43], we control for these two demographic characteristics. A dichotomy scale ("Woman versus man") was used for gender and a Likert scale with five categories (1: "18-25 years", 2: "26-35 years", 3: "36-45 years", 4: "46-55 years", and 5: "56 years or more") to measure age.

Analysis
To assess, the evidence of convergent validity and discriminant data were analyzed following two approximations. On the one hand, the correlations were analyzed using Pearson's r coefficient, between the different variables used. Positive correlations were considered evidence of convergent validity, while the correlations negative and the absence of correlation were considered evidence of discriminant validity. On the other hand, a confirmatory factor analysis (CFA) in the framework of structural equation modeling (SEM) was employed using EQ6 [44,45].
Hierarchical regression analyzes were performed using the SPSS statistical program (version 25) to evaluate linear and nonlinear relationships [46]. The regression analysis process consists of three steps: In the first step, the control variables (that is, sex and age) were entered to see their influence on the dependent variable. In the second step, the independent term (role overload) was included in order to test its linear relationship with the dependent variable (perceptions of intellectual engagement) in situations of role overload with medium and low levels. Finally, role overload was entered into the model as a linear overprint indicator (squared term). If the contribution of the squared term is statistically significant, the maximization hypothesis is rejected [47] and graphical representations are used to show support for fairness (symmetric) or asymmetric. The independent variables were focused on reducing possible multicollinearity problems [48].

Descriptive Analysis
Regarding the evidence of convergent validity, we found significant positive correlations between the items that make up the role overload scale (significant values between 0.72 and 0.83). The same occurs with the items that make up the intellectual engagement variable, all of them show significant values between 0.76 and 0.91.
In relation to the evidence of discriminant validity, negative and weak significant connections were found between the total score of role overload and intellectual engagement (r = −0.09).
The multi-group SEM of the hypothesized model fitted to the data well: Chi-Square = 8.065, df = 8, p-value = 0.43, CFI = 1.000, RMSEA = 0-003 (90% CI 0.000, 0.044). Figure 1 shows standardized parameters for the model [49].  Table 1 shows the means, standard deviations, and bivariate correlations between the variables. We observed a significant negative correlation between perceptions of role overload and intellectual engagement (r: −0.08; p: 0.03) and a significant positive correlation between age and intellectual engagement (r: 0.09; p: 0.02). The relationship between role overload and age was not significant.

Regressions Analysis
The results of the hierarchical regressions are presented in Table 2. The results indicated that the linear terms of role overload predicted a reduction in the values of intellectual engagement in the employees (β: −0.05, p < 0.05). On the contrary, the linear overprint (squared role overload) was also significant, but with a change in influence, which in this case will be positive (β: 0.06, p < 0.05). This shows variation in intellectual commitment, beyond the predictive capacity of the linear term and the control variables (∆R 2 : 0.03, p < 0.001). Therefore, our findings supported a curvilinear relationship between perceptions of role overload and intellectual engagement. Since the contribution of the squared terms was significant, the maximization hypothesis was rejected.

Asymmetric Curvilineal Analysis
The positive values of β in the final step show nonlinear relationships, represented by U-shapes (see Figure 2), which confirm the asymmetric curvilinear shape in the relationship between the variables. The greatest intellectual engagement is observed at the lowest levels of role overload. Intellectual participation scores decrease when perceptions approach the average levels in the perception of role overload. Lastly, intellectual engagement improves slightly when perceptions approach the area of role overload.

Discussion
This study aims to shed new light and learn more about role overload in the organizational context. This variable can show more than one role, with a positive influence on the intellectual engagement. Our results found in the present study support both hypotheses. In the first, it was proposed that workers subjected to work contexts with increasing levels of demands, such as role overload, will initially experience a reduction in their level of intellectual commitment to the organization (β: −0.05, p < 0.05). In the second hypothesis, it was proposed that the relationship will not be linear, but curvilinear, between role overload and intellectual engagement, generating an increase in the latter variable at the highest levels of perception of overload (β: 0.06, p < 0.05).
These results confirm the U-shaped curvilinear relationship [30,33,34]. At the lowest values of overload, this variable acts as a hindrance demand, producing the reduction of intellectual engagement. However, at high values of role overload, this acts as a challenging demand, producing an increase in the response of this type of engagement. This is a finding that provides a new point of view with respect to previous studies that indicate that work demands such as role overload, conceptualized as role overload, produce an inverted U relationship where the response as a challenge occurs in the initial values of the perception of demand [24].
In the following sections, we describe the theoretical and practical implications, limitations, and opportunities for future research.

Theoretical Implications
The results shown in this study have three theoretical implications. First, they clarify the conceptualization of job demands as an obstacle or challenge within the JD-R model [9,[13][14][15]. This model proposes that demands that act as obstacles interfere with the person's ability to achieve their goals [14,20], while challenging demands promote personal growth and are perceived as work experiences rewarding. [14,15]. Expanding the demand concepts of this model and supporting the results of Tays et al. [30], Taris et al. [33], or Baumeiste [34], and recently Stroe et al. [48] and Horan [49], the present work shows that the role of a demand is not static but fluctuates between obstacle and challenge depending on the level at which it is perceived by the employee.
The second theoretical implication of this work supports, from this fluctuation, the role as an impeding demand of those that are perceived as "excessive". In this case, role overload is conceptualized as an excess of workload, so low values of this variable cause deterioration of intellectual engagement. These results are in line with the resource conservation theory [36] and with the findings of the studies that are supported by the Yerkes-Dodson theory [32], which affirms that, when a demand requires a high expenditure of resources to be faced, it will have a negative effect on the emotional state of the employee [24,31,50,51].
The third theoretical implication of the present work is linked to the increase of intellectual engagement in the highest values of role overload. The results found in this regard suggest that the response to job demands occurs in different stages. Previously, authors such as Chen and Wei [52] proposed that the increase in role overload initially produces an increase in positive emotional responses in employees, but when they perceive this burden as excessive, the influence is reversed reducing their level of Engagement. The confirmation of the second hypothesis of the present work shows how, in the face of high perceptions of overload, employees once again modify their response and increase intellectual engagement again, although with less intensity. This result supports the hypotheses put forward by Seers, McGee, Serey and Graen [38] and Bakker et al. [39], and shows that in highly stressful conditions, people increase their positive emotional response again, not so much to reduce demand, but the personal consequences that arise from it.

Practical Implications
The findings of this study have practical implications for the management of people that we can group into three levels: (1) Analysis of tasks and job design that adapt the workload to the capacities of the employees; (2) implementation of strategies to increase the personal and organizational resources of employees, especially in those positions that face high levels of work overload maintained over time; and the (3) reduction of personal consequences derived from the existence of highly job demanding contexts, as recommended by the Sustainably Development Goals reports prepared by the United Nations [53]: (a) Employees are frequently subject to role overload demands, which easily overwhelm their resources and have negative implications such as sleep disturbances, development of emotional disorders, and deterioration of physical health [7]. The results of this study highlight the importance of proper job design, where employees are not exposed to a constant role overload that they cannot cope with. If the job design is adequate, workers will have the necessary resources to cope with the role overload, which will generate new positive resources in them, such as increased selfperception of effectiveness. Furthermore, this increase in resources will allow them to face unpredictable role overload increases, due to specific conditions of need, with greater guarantees. (b) The results found indicate the convenience of designing strategies to increase the personal and organizational resources of employees. Giving employees more resources can make it easier to cope with the demands of each position, which can mean that the employee can withstand greater levels of role overload before it becomes a hindrance demand. (c) Finally, the asymmetric hypothesis advises the development of coping skills of employees on a personal level in the face of stress. We have seen that employees return to give a positive emotional response to a hindrance demand when it is at very high values. This has been considered by authors as a protective response, which does not seek to face the demand, but rather to reduce the negative consequences for the individual [38,39]. In this situation, organizations must provide their employees with resources for their own protection against demand, such as techniques of emotional self-control or mindfulness, while implementing policies to reduce the exposure of employees to said demand.

Limitations and Future Directions
This study has three main limitations. First, it was conducted within a specific organization, so the results cannot be extrapolated to other work contexts. Future studies could expand the sample to include employees of other public or private organizations.
Second, there are limitations in that it only uses self-report questionnaires. Combining this with other types of approaches could have provided complementary data. Future research may include the use of qualitative methods [54]; such as in-depth interviews, or other tools such as observation or the use of objective data on the behavior of employees in the organization (for example, productivity or number of sick leave).
Finally, this is a cross-sectional study with only one antecedent and one consequent variable. Therefore, it would be useful to carry out longitudinal studies that allow a more in-depth analysis of the evolution and causality of the variables studied. It would be very interesting for future research to explore certain resources in employees or in the organization that may affect the relationship proposed in this article. These factors will allow us to better predict the response of employees' intellectual engagement in the face of work overload.

Conclusions
The present study contributes to deepening the knowledge of the different roles that labor demands can assume in the organizational context. The finding of nonlinear relation-ships between role overload and intellectual engagement presents a novel hypothesis on which to continue working: the appearance of a demand in the work context will cause it to become a challenge and will initiate a positive response in the employee. If the employee sees that he is unable to respond to the demand with the resources available to him, this will become a hindrance demand, and will generate negative consequences for the employee and the organization. However, if the demand level continues to increase, the employee will once again give a positive response to the demand, not so much to face it, but to protect himself before the negative consequences of it.