Introduction

Psychosocial risk factors as determinants of health at work are present in all work situations related to the organization, content, and performance of tasks [1]. These risk factors can affect the worker’s physical, psychological, and social well-being [1,2].

There is evidence that the motivational attention pattern and affective states are related to the processing of emotionally positive or negative content [3]. Moreover, training systems that include cognitive-preventive content work best when the mechanisms of such knowledge construction are based on neuroscience [4].

In particular, preventive interventions are a vehicle for critical thinking and learning when the focus of attention contains few elements or fragments consisting of words or digits, visual-printed objects, and abstract ideas [5]. An example of preventive interventions are the ones promoted by training with a neurolearning approach.

The literature points out that the processing of emotional content in the occupational and clinical field also works best when the focus of attention, or when the learning system, is supported by neural feedback [4,5,6]. Cognitive-emotional sentences processed according to neural elements generate positive changes in occupational health and safety inspections [7,8], as content processing responds to semantics and visual feedback of pleasant, unpleasant, and neutral emotional texts [9,10].

This also extends to deterrent and persuasive phrases as psychological stimuli that modulate attention and emotional and cognitive reactions as descriptive or emotive meanings. Specifically, deterrent phrases have greater motivational significance because they allow for more elaborate processing of severity or deterrence content than those used for adaptive protection, given that they are very sensitive and have a greater impact on content with emotional connotations. For example: "mandatory inspections are carried out" or "fines are imposed" [9,11]. Meanwhile, persuasive sentences are more concerned with assertive language, which also influences feelings and persuades people to change their behavior by alerting them to future actions or reorienting their system of interests. For example: "we would appreciate delegating to the occupational health representative" [8,11].

Regarding the neuropsychological frame of reference, visual working memory and the location of objects in the environment related to the subject of the study are usually processed in the occipital lobes [12]. The left frontal lobe processes working memory, executive planning, immediate attention, concentration, and executive planning [12].

Moreover, a significant body of data indicates that brain neural activities reflect the focus of attention and thinking functions, cognitive processing, learning, and problem-solving tasks [8,13,14]. While the right hemisphere is responsible for episodic memory functions, the frontal poles are responsible for non-verbal memory functions, attentional judgment, and spatial awareness [13,14].

It is also known that content processed from graphic stimuli after visual feedback is globally transformed in procedural working memory [14,15] and that internal verbalization is achieved by stimulating Broca’s area [8,14]. Thus, the neural features resulting from the simultaneous application of isometric concentration and symbolic thinking are induced by activating the frontal lobe [15,16]. Then, sensory systems and visual feedback are further integrated due to the generic activation of procedural working memory, and the motor performance obtained by combining digital-therapeutic relaxation and isometric concentration is achieved [5,14,16].

According to the available literature, promoting occupational safety culture in Latin America through preventive interventions in the work environment through neurolearning-based training is very limited [17]. However, worker conditions are part of the social determinants of occupational health, which require improved preventive practices within the framework of international guidelines [18].

In the Peruvian organizational context, the workplace must address psychosocial factors with the same intensity as physical, chemical, biological, or ergonomic risks. This is because these factors have become global problems that affect all occupational groups with a substantial impact on health and work performance [17,18]. Therefore, in occupational health and safety management, the promotion of preventive interventions constitutes a competitive advantage, as it is considered a parameter of corporate social responsibility that favors the promotion of learning and a culture of safety at work. The effects of these preventive actions are favorable when these interventions are developed within the same work environment [2,7,19].

The study aims to compare the degree of processing of cognitive-emotional content with the insertion of deterrent phrases (group A) versus the insertion of persuasive phrases (group B) as an effect of training with a neurolearning approach to psychosocial health at work in a group of professionals with health insurance in the Peruvian Amazon.

Methods

Design

An experimental design with pre-test and post-test control groups was used (A/B: O1──X──O2; C: O1── -.- ──O2), where the response variable is the processing of content on psychosocial occupational health regulations. The unit of outcome measurement is the number of iconic features (Figure 1).

Figure 1

CONSORT-SPI flow chart.

CONSORT-SPI: Consolidated standards of reporting Trials-Social and Psychological Intervention.
PHW: Psicosocial helath at work.
Source: Prepared by the authors of this study.

Full size

Iconic features are the number of graphically processed elements within a mental map related to psychosocial occupational health regulations. They are processed interchangeably as semantic categories (in form and location), symbols or figures, words or phrases, acronyms or codes, and through other combined semiotic uses [20,21]. The processing of the above content was compared between group A, with 25 deterrent phrases related to compliance with psychosocial health at work regulations, and group B, with 25 persuasive phrases. The control group C was exposed to neither deterrent nor persuasive sentences (Figure 1). The intervention variable is training with a neurolearning approach, targeting groups A and B. The three neuropsychological stimulation techniques used were:

1. Self-relaxing digital therapy.

2. Visual reconstruction.

3. Isometric concentration, accompanied by daily sports practice.

Control group C received conventional training (Figure 1).

The purpose of the control group was to guarantee the credibility of results in the pre-and post-intervention assessment, to evaluate the influence of external variables to ensure that the performance in post-test content processing is the effect of the training, and that the randomization of participants in the intervention groups (A, B) concerning the control group are similar in number of participants.

Participants

Participants for our study were recruited from a training cycle on occupational safety and health, promoted by a professional group, and developed in a university institution in the Peruvian Amazon. The study took place between December 2018 and January 2019.

The sample size was calculated using the G*Power 3.1.9.2 calculator, considering a minimum effect of results (η2 = 0.06 to 0.13), a significance level of 0.05% and a power of 95%, the result of which was n = 54/3; that is, 18 participants per group. The sample size determined for this study is consistent with other work [21].

The 48 selected attendees were randomly assigned to intervention groups (A and B) and control group C (16 subjects for each group). A random number table was used to distribute the subjects in a 1:1 ratio (Figure 1). The groups' balancing occurred post-randomization using sex, profession, and position of the participants [22].

The inclusion criteria were as follows:

1. Voluntary and anonymous participation.

2. Being affiliated with a health insurance scheme.

3. Be an engineer or graduate in the food industry.

4. Hold the manager or delegate of occupational health and safety position in a microenterprise located in the Peruvian Amazon.

5. Signing informed consent form.

The exclusion criteria were

1. Failure to complete the mind map.

2. Withdrawal from the study.

3. Lacking health insurance.

4. Be 60 years of age or older.

5. Have training other than in the food sector.

The masking was single-blind, as the participants were unaware of both the group to which they were assigned and the insertion of the deterrent or persuasive phrases in the text. This masking was achieved by presenting the same text and format on psychosocial health at work in all three groups, both pre-test and post-test. Researchers were unblinded to group selection.

Training with a neurolearning approach

Before the pre-test, all participants were trained in constructing content using a mind map. The mind map is an appropriate technique for representing graphic analysis results based on the comprehension and summary of a text [23] because it increases interest in hierarchical learning of related concepts as it favors the identification of variables with their causal relationships due to its multidisciplinary approach [24].

Two weeks before the end of the conventional training cycle, the members of the intervention groups (A and B) were trained through behavioral modeling on the application sequences of the neuropsychological stimulation techniques with a neurolearning approach (digit-therapeutic self-relaxation, visual feedback, and isometric concentration). These techniques are still undergoing empirical verification [21]. However, some of their benefits assumed in other studies are:

1. Therapeutic digital self-relaxation technique: It has psychophysiological effects for relaxing the autonomic nervous system, allowing for improved cognitive processes.

2. Visual feedback technique (maximum 3.5 minutes): Helps to improve experiential learning. Better results are obtained when applied in small groups and associated with other neuropsychological stimulation techniques [25,26,27].

3. Isometric concentration technique: Among other benefits, it stimulates working memory, muscle relaxation, and balance awareness, favoring kinaesthetic awakening and the fixation of images constructed in the mental map [21].

During the post-test, participants were admitted from 7:30 to 9:30 in the morning, considering that the highest performance of cognitive activities, such as short-term memory [26], occurs in the early morning hours [28,29]. Sports practice was suggested following Petracovschi & Gombos [30] for six weeks (minimum 30 minutes daily and before attending training sessions). Participants were placed approximately 2.5 meters apart and in groups of a maximum of eight people per room (a total of six rooms). After the semiotic content analysis of the mind maps, the frequency of iconic features and socio-occupational data were recorded. The application time per participant was between 28 and 40 minutes.

The conventional training, including the neurolearning approach training, in groups A and B was of equal duration (18 hours distributed over six weeks). While the conventional training in group C was 15 hours (spread over five weeks), this group did not receive neurolearning approach training.

Data collection and analysis

Knowing that the primary source of information is the mental map generated by the participants, the iconic features were quantified and recorded on a card previously validated by five experts (organizational psychologists and labor lawyers), whose results of the statistical analysis of concordance in four endpoints were adequate according to Kendall’s W test (W = 0.800; gl = 3; p = 0.007). The agreement was significantly different from zero (p < 0.05) [31].

Activities for statistical analysis followed the standards of the Consolidated Standards of Reporting Trials for social and psychological interventions, CONSORT-SPI, with statistical techniques applied sequentially according to the assessment pathways of the response variable:

1. The normality of the data distribution was evaluated using the Shapiro-Wilk test to verify the similarity of pre-test responses in the three groups (A, B, C). As results followed normal distribution, ANOVA was applied for independent groups (Annex).

2. The normality of data distribution in both stages was determined to assess that the results of the control group’s response variable remained unchanged in both pre-test and post-test. The difference in results was then calculated using the Wilcoxon test for related groups, N < 30 (Annex).

3. The normality of data distribution was also assessed to verify that the response variables in the intervention groups (A and B) are significantly different between post-test and pre-test. Then, as the result was inadequate (p < 0.05) and N < 30, the Wilcoxon test for related groups was applied (Annex).

4. As data followed a non-normal distribution (p < 0.002), the Kruskal-Wallis test for independent groups was applied to assess that the post-test response variable was significantly different between the two intervention groups and the control group (Annex).

5. The Kruskal-Wallis test was applied to compare the results of the response variable between pre-test and post-test between groups because it did not meet ANOVA assumptions. The effect sizes were calculated for the outcomes, ε², and the post-hoc pre-test match (rpb).

The data were processed using STATA v.15.

Results

Participant characteristics

Of the total 48 participants aged between 22 and 36 years (mean: 23.0 ± 2.08 years), 64.6% (31/48) were males distributed equally in groups B and C with 32.2% (10/31) each, and in females with 35.3% (6/17) each also in groups B and C. According to the profession, group A had an equal distribution of professionals and technicians in the food sector, with 50% (8/16) each. Group B had 56.2% (9/16) professionals, while Group C had 56.2% technicians. According to the position in the microenterprise, 56% of managers were distributed in Group A, 62.5% in Group B and in Group C. Both managers and occupational health and safety representatives were equally distributed with 50% (8/16) each (Table 1).

The normality of data distribution by groups was adequate in both pre-test and post-test (p > 0.050), except in group B, which was not adequate, p = 0.002 (Table 1). In the pre-test stage, the homocedasticity, i.e., the results of the response variable and its adequate normality of data distribution (Table 1), was also adequate (p = 0.115). The results presented in Table 2 are supported, as there are no significant differences in the processing of psychosocial health regulations at work between the three groups (F = 1.683, p = 0.197).

In the control group, the results of processing content on psychosocial health at work in both the post-test and pre-test remained similar (p = 0.667). This ensures that the changes in the results in the intervention groups ("A" and "B") are acceptable in the sense that the results obtained in the post-test are significantly different from those obtained in the pre-test, p = 0.001 (Table 3).

Results also confirmed that the processing of psychosocial health content at work in the post-test is significantly different between the intervention groups (A and B), and in turn, concerning the control group’s results, p = 0.001 (Table 4).

The processing of the response variable was significantly different between post-test and pre-test in group A versus group B (p = 0.001), with a large effect size in both groups (A, ε² = 0.758; B, ε² = 0.760) (Table 5). It can be distinguished that the processing of content with deterrent sentence insertion (group A) on psychosocial health at work is significantly different (median: 46 versus 41.5) and with larger effects compared with group B (p = 0.001, rpb = 0.949) (Table 5).

Likewise, Group A had a larger improvement of iconic traits related to psychosocial health at work, raging between 22 (20 to 42) to 26 (20 to 48) compared with Group B, where the increase was smaller: ranging from 14 (19 to 33) to 24 (19 to 43) iconic traits. The predominant score range with optimal performance in group A was between 45 and 48 iconic traits for 50% (8/16) of the participants.