To the best of our knowledge, this was the first study to examine autonomic changes in young men with HIA and LIA using 24 hour HRV monitoring. While no difference was observed in the HRV band power change (minmum-maximum) between the HIA and LIA groups, we found a significant difference in the daily average of HF and SDNN. However, no significant difference was found in HF during the IGT or when the participants played their preferred games.
The results of previous reports on HF at rest showed that HF decreased with high depression (Blood et al., 2015) or Internet dependence, which was the opposite of the present results. Participants in this study did not demonstrate the severity of Internet addiction observed in previous studies(Chang et al.,2015; Hesieh et al., 2016; Hong et al., 2018; Kim et al., 2016; Lee et al., 2018; Lin et al., 2014; Moretta et al., 2018 ). The present study was conducted on healthy subjects classified using the IAT, and the low level of severity was likely the reason that no significant differences were found, except for HF. HRV is related to frontal lobe function, and moderate game use has been reported to have a positive effect on it (Blood et al., 2015; Smirni et al., 2021).
The clinical usefulness of HRV remains poorly understood, and whether short term or 24-hour measurements are better is unclear. For instance, Fei et al. (1996) demonstrated that the SDNN of a 5-minute ECG recording could predict the mortality of patients with myocardial infarction, but was inferior to long- term HRV indices. In contrast, Bigger et al. (1996) compared the power spectrum of short ECG recordings with 24-hour long-term HRV, and concluded that both HRV analyses were excellent predictors of mortality in patients with myocardial infarction.
At the early stage of Internet addiction, psychometric testing appears superior to HRV monitoring for evaluating the psychological factors contributing to this aberrant behavior. The detection of Internet addiction could be improved by using multimodal biosignals such as EEG and galvanic skin responses (Kim et al., 2019). However no significant results were found compared with baseline heart rate variability (Chiu et al., 2023) .
Hong et al. (Hong et al., 2018) also observed significant reductions in HF power during periods of high attention and last 5 minute of a task. The time course of the game may influence HRV. However, we did not find time course changes in HF between the groups during the IGT.
When the overall results were classified by age, significant differences were observed in IGT scores and responses to HF between high-school students and young adults. In other words, IGT scores were significantly higher in young adults than in high-school students, and responses to HF were greater among young adults than among high school students. This was consistent with the findings of Beitz et al.’s (2014), who found that adolescents had poor decision-making processes and low IGT results. This was also consistent with the results of Forte G et al.’s (2021), who reported that good decision-makers had a high vagal tone during the IGT. As shown in Table 2, a significant difference was observed in BMI between the groups. However, Antelmi et al. (2004) found no significant difference in HRV between normal, overweight, and obese groups of men and women. Therefore, it is reasonable to limit the time and content of Internet use for high school students who are maturing in their decision-making process, as has been conventionally established. According to a recent review, a small but significant relationship exists between HRV and top-down self- regulation, which is thought to be stronger in older individuals (Laborde et al., 2017); the observed differences in HRV between the age groups in our study suggest such a relationship.
A positive correlation was found between HF, IAT and BDI-II among high school students. In other words, higher tendency toward Internet addiction was associated with higher levels of depression, and higher autonomic nervous system function. Although this result seems to contradict conventional findings, HF may not decrease in the early stages of Internet addiction and depression. Selye's (1976) presented a three-stage stress response model: in Stage 1, the sympathetic nervous system is activated during the warning response phase: in Stage 2, the body attempts to respond to stress in the resistance phase, and the vagus nervous system is activated to normalize the physiological response, while blood glucose, cortisol, and adrenaline levels remained high. : In Stage 3, the body loses its resistance. Therefore, in our case, the participants were in Stage 2, and demonstrating increased HF.
This study had several limitations. First, the small number of participants may have prevented the detection of statistically significant differences. Thus, future large-scale studies are warranted. Second, we did not monitor respiratory changes during game play, and differences in respiratory responses could have influenced the HF band of the HRV spectrum. However, a recent study found no significant group difference in respiration during gaming tasks (Chang et al., 2015). Third, we did not consider specific details of participants’ Internet use such as excessive social networking, online shopping, gambling, or pornography use.