From interpersonal empathy to digital empathy

The rapid adoption of social and mobile technologies altered, as many other aspects of contemporary culture, society’s communication patterns and, somehow, disrupted the expression of empathy (Terry & Cain, 2016). The authors referred to a new form of empathy, the digital one, that they define as the “traditional empathic characteristics such as concern and caring for others expressed through computer-mediated communication (CMC)” (Terry & Cain, 2016: 1).

If the importance and impact of technology evolve as expected, digital empathy will become an even more critical issue in all aspects of life, including education. When technology meets the online educational space, emotions become mediated. According to Ahmed (2022: 76), one of the most relevant aspects to be discussed here is “how digital tools spatialize bodies and emotions of students and teachers”. Consequently, digital empathy arises as a way to connect emotions and pedagogy. Once empathy has a strong correlation with positive educational outcomes and it is a fundamental part of the student-teacher relationship, the technological advances, the increasing digitization of education and the consequent decrease of empathy may be cause for concern.

Empathy and neuroscience

In 1992, three Italian researchers found that monkeys, when watching them eating, showed the same neural activity. Deepening this insightful incident, they discovered the “mirror neurons” – nerve cells that enable humans to understand other’s experiences by undergoing a kind of involuntary “echo” while observing one another’s behavior (Gerdes et al., 2011). This discovery marked the beginning of the empathy research within neurosciences. In 2004, Decety and Jackson (2004) observed specific brain activity related to empathy, being able to identify the neural mechanisms that mediate empathy in the brain.

Although D’Ambrosio et al. (2009) assumed that there is consensus on the two dimensions of empathy (affective and cognitive), Carré et al. (2013) stated that recent neurophysiological studies have allowed researchers to define empathy as underpinned by three components: cognitive empathy, emotional contagion, and emotional disconnection. According to the authors, emotional contagion involves automatic processes that allow immediate evaluation of the emotions’ nature: positive or negative, pleasant, or aversive, and involves subcortical structures such as the limbic lobe. The cognitive empathy involves activations of the insular cortex, the ventromedial prefrontal cortex and medial PFC. Finally, emotional disconnection is related to executive functions based on the orbitofrontal cortex, medial and dorsolateral prefrontal cortex, and the anterior cingulate cortex (Carré et al., 2013: 681). Once modern neuroscience reveals the impact of positive emotions on cognitive processes during learning activities, this fact emphasizes the need to develop digital empathy strategies in education (Li et al., 2020: 227).

Empathy and education

Authenticity and empathy are qualities that routinely appear on an attribute list of a “good” teacher (Bialystok & Kukar, 2018) and there are numerous studies relating empathy with positive findings in prosocial behaviors such as cooperation, sharing, donating, and other altruistic acts (Batson et al., 1981), all relevant in the learning process. The classes transfer to an online environment broke many of the interpersonal connections existing in the physical classroom. As known, traditional face-to-face communication of personal thoughts and attitudes widely differs from the digital one (Terry & Cain, 2016). If learning is basically a social process (Swan & Shea, 2005) and empathy is a central concept in the context of social interactions, it is of utmost importance to understand the interaction between students and professors in online classes where communication is essentially computer-mediated (Lowenthal, 2010).

During the pandemic, some higher education institutions innovated their teaching systems by creating and implementing diverse academic models focusing on positive emotions and social inclusion in a period when many young people may have been experiencing a feeling of alienation. Okoye et al. (2021) mentioned two interesting online educational models: HyFlex+ TEC (a hybrid and innovative program) and MFD+ (a flexible digital model), highlighting the impacts on emotional health of the latter.

The preocupation with the impact of digital education on students’ performance and emotions is not new. In 2000, Garrison et al. (2000), while proposing a conceptual tool for CMC in education, presented the Community of Inquiry (CoI), developing the concept of successful educational experience based on three relevant aspects: cognitive presence, social presence, and teaching presence. This framework contextualises the relationship between education and empathy in the digital environment. The nature of online communication allows individuals to be physically invisible to others (cameras off), which invalidates the crucial non-verbal communications: facial expression and eye contact (Nelson-Jones, 2005), or possible awareness of the person's physical reaction to another person(s). As much of traditional face-to-face communication tends to be non-verbal, without cues like tone of voice or body language, digital interactions lack this essential element of understanding (Terry & Cain, 2016).

Another important issue potentially lost in online education is the emotional contagion, that is developed during the preverbal period and constitutes the first step in the empathic functioning (Carré et al., 2013). Although researchers agree on the positive role of empathy in interpersonal relationships (Stephan & Finlay, 1999), there is evidence that CMC altered interpersonal dynamics (Konrath et al., 2011) and it is currently acknowledged that empathy is influenced by age, gender, and other attributes of the individual, including situational context (Feshbach & Feshbach, 2009). Therefore, we hypothesize:

• H1. Students spending more time in online education display a lower level of empathy.

• H.1.1. Students spending more time in online education display a lower level of cognitive empathy.

• H1.2. Students spending more time in online education display a lower level of emotional contagion.

• H1.3. Students spending more time in online education display a higher level of emotional disconnection.

• H2: The level of empathy varies according to gender in online education.

• H2.1 Female students are more likely to display a higher level of empathy than male students when attending online education.

• H3: Students with low level of empathy are more likely to switch off their camera during online education.

• H4: During online education, students with a high level of empathy are more likely to need the emotional support of their colleagues.

Participants

The main criterion of respondents’ selection was the field of study: Communication. The exact number of students enrolled every year in Portugal and Romania in Communication Studies is difficult to determine, since in both countries the only official data refer to the maximum number of places available in those study programs, not to the actual number of enrolled students. Given this limitation, a non-probability sampling technique was applied to obtain a convenient sample (Bryman, 2012; Kalton, 2020), which included 570 Portuguese and 515 Romanian students.

The sample (n=1,085) consisted of volunteers contacted personally via institutional e-mails, social media groups, and with posters with a QR code and a presentation of the research, displayed in visible places on the premises of Communication Departments/faculties in both countries.

All participants were informed about the purpose of the research and were free to withdraw from the survey at any time. The respondents had to meet the following requirements: be enrolled in Communication programs in Portuguese or Romanian universities (at any level); have a fluent level of English (language of the questionnaire); be at least 18 years old, and freely agree to participate in the survey through an expressed informed consent.

Regarding the age distribution, more than 80% of the participants were young adults, 18 to 23 years old, with most respondents aged 18-20. What is predominant in Portugal with 55.26% are Bachelor students aged 18-20 in comparison with 27.72% aged 21-23. In Romania, there is a visible age balance between the first two categories, 46.01% representing the participants aged 21-23, and 41.17% aged 18-20.

In terms of gender (Table 1), young women represent 73.36%. This can be explained by a majority among the students enrolled in both countries in Communication programs, as this field offers many opportunities for women at European level (Ross & Padovani, 2017).

Data collection procedures and tools

The questionnaire was anonymous and did not collect names, e-mail addresses or geo-locations. Data collection was conducted through a self-administered online questionnaire (Bryman, 2012), by using the link provided by the researchers or by accessing the QR code included in the poster. The questionnaire consisted of four sections: (1) demographic data regarding the age group, country of origin, gender; (2) exposure to the usage of digital technologies (number of hours spent with digital technology) and to online education (the number of semesters with online education); (3) level of empathy (see “infra”); (4) online education self-perception involving the use of webcams and self-evaluated feelings regarding online education (“online education positively impacts my life”, “during online teaching sessions, I find it difficult to interact with my classmates”, “after 2 years of pandemic, students are tired of being nice in online classes”, “after 2 years of pandemic, I feel more disconnected emotionally”).

The third section - measuring empathy - consisted of a 20-item scale entitled Basic Empathy Scale (BES), designed by Jolliffe and Farrington (2006). This scale was tested on British adolescents, and the results were interpreted using two factors: affective and cognitive empathy, which may overlap, but show a degree of differentiation (Jolliffe & Farrington, 2006: 602). In 2013, Carré et al. validated the BES for adults, and proposed a three-factor model of analysis: emotional contagion, emotional disconnection, and cognitive empathy. Emotional contagion is a form of synchronizing one’s own postures and expressions with those of the persons around (Hatfield et al., 1994). Emotional disconnection refers to “a mechanism of disconnection from emotion that protects individuals from excessive emotions” (Carré et al., 2013: 686), whereas cognitive empathy means a manner of immediately detaching oneself from the other’s emotions, so as not to suffer. For Gerdes et al. (2011), the coherence and combination of the 3-factor model of empathy can be justified by neuropsychological processes as well. This section made use of the BES proposed by Jolliffe and Farrington (2006) and led to the interpretation of the results using the 3-factor model of Carré et al. (2013) that excluded one item from BES, but to facilitate comparison between this study and previous similar ones, the mentioned item (4) was kept in this research.

Following both survey designs (Jolliffe & Farrington, 2006; Carré et al., 2013), the participants rated each sentence on a 5-point Likert type scale (-2 Strongly Disagree, -1 Disagree, 0 Neither Agree nor Disagree, 1 Agree, 2 Strongly Agree). To uniformly interpret the results, the Likert scale was re-coded as follows: 1 Strongly Disagree, 2 Disagree, 3 Neither Agree nor Disagree, 4 Agree, 5 Strongly Agree. In the original BES, seven items were reversed (to be empathic the respondent had to disagree with the item), therefore the scale was also reversed in the case of these 7 items to provide a uniform interpretation of the results. Therefore, the minimum level of empathy is expressed by 20 points, and the maximum by 100 points (Carré et al., 2013: 682). The statistical analysis was performed using free softwares: PSPP version 1.6.2; R version 4.1 (R Core Team, 2021); R statistical packages (Fox & Weisberg, 2020; Lenth, 2020); Jamovi Version 2.3.

Digital technology and online education

91.61% of the respondents declared that they spent more than 4h/day using digital devices, and 36.68% spent more than 8 hours daily using computers, tablets, or smartphones. The variation between the two countries is below 3%. For example, 56.32% of the Portuguese students considered they spent between 4 and 8h/day, similar to 53.40% of Romanians. The high exposure to digital technology representing at least 8h/day unfolds similar self-estimated results, i.e., 35.09% in Portugal and 38.4% in Romania.

Regarding participation in online education, there are significant differences: 31.93% of Portuguese participants declared that they attended 2 semesters of online courses, compared to 16.12% of Romanians, and 9.82% of Portuguese attended 4 online semesters, compared to 32.63% of Romanians. Both Portuguese and Romanian students self-assessed that they had studied 3 online semesters, representing almost the same percentage (29.82% in Portugal and 29.13% in Romania). Despite the pandemic context, 7% of total respondents considered they did not attend online classes, whereas, conversely, 1.57% self-evaluated their online studies as over 5 semesters.

The level of empathy

The level of empathy - calculated by using the total number of points for the 20-item BES as described “supra” - varies between 43 and 100 points (Table 2).

Most participants (920, i.e., 84.79%) displayed a medium level of empathy between 61 and 90 points. The average level of empathy in Portugal is 76.20 (SD=9.69) and in Romania is 75.51 (SD=10.57).

Regarding gender, women scored higher than men, irrespective of the country of origin. For the total empathy scale women had an average of 77.63 (SD=10.11), and men 70.72 (SD=9.03). The average score of Portuguese women was 78.23 (SD=9.18), and men 71.29 (SD=9.15). In their turn, Romanian women had an average score of 77.04 (SD=10.53), and men 70.16 (SD=8.85).

The data on emotional contagion (the interval is 6-30) displayed no differences between countries, but they show significant differences regarding gender. In Portugal, the average score of emotional contagion (women and men) is 19.97 (SD=4.29), while in Romania it is 19.84 (SD=4.83). In Portugal, women’s average score for emotional contagion is 20.91 (SD=3.93), and men's 17.69 (SD=4.26). In Romania, women's average score is 20.57 (SD=4.59), and men's 17.27 (SD=4.81). Women score higher than men in both countries.

The data on emotional disconnection (the interval is 6-30) also displayed no significant differences in terms of country of origin but show differences regarding gender. In Portugal, the average score for emotional disconnection (women and men) is 24.31 (SD=3.98), while in Romania it is 23.17 (SD=4.32). The difference between women and men scores within the same parameters (around 3 points out of 30). In Portugal, women's average score is 24.98 (SD=3.67), and men’s is 22.71 (SD=4.24). In Romania, women's average score is 23.89 (SD=3.96), men's 20.68 (SD=4.59).

The impact of online education on students’ lives

Although there is a difference between Portuguese and Romanian students, 63.50% of respondents estimated that online education had a positive influence in their life. 68.42% of the Portuguese participants “strongly agree” and “agree”, compared to 58.06% of the Romanians. Only 7.37% considered that online education did not have a positive impact in their life.

During online classes, 47.85% of the participants believed that they had no difficulties interacting with colleagues. In Romania, 51.85% of the students self-assessed that online education did not make communication with classmates difficult, an opinion similar for 43.86% in Portugal.

The level of students’ interaction by using a web camera

The percentage of participants who “agree” or “strongly agree” with a difficult interaction during online classes is 25.53. The difference between Portuguese and Romanian students who self-evaluated very low interaction with their fellows is less than 2%. Similarly, 19.65% of the Portuguese admitted that the interaction with their classmates during online teaching was difficult, while the Romanians self-evaluated with 16.12%. The total percentage of “neither agree nor disagree” assessment is 28.82%. More than half of the Romanian students did not agree with the difficulty of interaction with classmates, compared to 43.86% of the Portuguese. The difference of 8.99% between the Romanians and the Portuguese students is significant.

75.02% of the respondents participated in online classes with the camera off, without differences between countries. Regarding the attendance with the camera on, there is a difference between countries: 26.66% of Portuguese, compared to 19.41% of Romanians.

The level of empathy “fatigue”

44.52% of the Portuguese and Romanian participants “strongly agreed” or “agreed” with the idea that it is difficult to be nice in online classes (Table 3). This feeling could be associated with an “empathy fatigue” or empathy blocking. Even so, 15.61% of the Portuguese declared they are still willing to communicate nicely in online classes, while in Romania the percentage is 18.25. Only 10.97% disagreed with the possibility of getting tired of being nice in online classes, after 2 years of the pandemic.

Self-assessed emotional disconnection after pandemic

In this research, 41.47% of the respondents self-evaluated as emotionally disconnected (Table 4). The percentage of Romanians stating that they “strongly agree” and “agree” with emotional disconnection after 2 years of the pandemic is 45.44, while among the Portuguese participants it is 37.90. In terms of disagreement and strong disagreement with emotional disconnection, the Portuguese self-assessed with 42.81%, compared to 35.73% of the Romanians. The percentage of Portuguese and Romanian students who neither agree nor disagree with this emotional disconnection is 19.08, well balanced between both countries.

Gender, empathy, and technology

The results of the data analysis from both countries show that there is a correlation between the level of empathy, and gender (Table 5). A weak correlation can also be observed between the empathy and the hours spent online and number of years in higher education.

The ANOVA test was run to investigate the impact of factors on digital empathy. With a 0.05 level of significance, gender, the number of years in higher education, and the number of hours spent online seem to impact empathy (p-values<0.05). To identify the categories that impact empathy differently, Tukey contrast tests were run, and the conclusions are the following: female empathy is higher than men’s (mean difference=7.2); men’s empathy is lower than others (mean difference= -11.7) and students with less than 3 years of study in higher education display a lower level of empathy (mean difference= -2.8). Furthermore, the ANOVA tests show that emotional contagion is mainly impacted by exposure to digital technology.

The participants with a higher level of empathy are more likely to need emotional support during online education (p< .001) (Figure 1). Tukey contrast tests were run, and the conclusions were that those who most agree with the need for emotional support (4 and 5) have greater empathy (for example, mean difference between 1 and 5 is -5.606, and between 2 and 5 is -5.394).