Impact of information on public opinion reversal—An agent based model

Modeling Public Opinion Reversal Process with the Considerations of External Intervention Information and Individual Internal Characteristics

“…For instance, Huang et al [ 13 ] introduced a new model, which took the relationship between individual cognitive bias and their corresponding choice behavior into account, and pointed out that cognitive bias and the change ratio of choice behavior dominated the probability of public opinion reversal. Zhu and Hu [ 14 ] analyzed information strength, time of releasing information and different types of counter information through a large number of simulation experiments, to find out the principles of public opinion reversal. Schmidt et al [ 15 ] Evaluated Facebook users’ opinion polarization on the topic of vaccination, and made the conclusion that the introduction of dissenting information into a sub-group was disregarded, and could produce a backfire effect, thus reinforcing the pre-existing opinions within the sub-group.…”

“…Since the literature [ 14 ] has conducted in-depth discussions on the influence of external information on public opinion reversal, this paper only briefly expounds the influence of external intervention information. The information in the public opinion field usually has a certain extent of “untruth”, so the truth of the event often needs various evidences.…”

“…Figure 8 represents the simulated group opinion evolution diagram and group attitude distribution histogram based on the literature [ 14 ]. As can be seen from Figure 8 , under the condition that other simulation conditions are consistent, when I 1 = −0.6, the group attitude is under the influence of negative information, the attitudes of most individuals are between −0.6~0, the group attitude is relatively negative, but it has not formed an obvious negative polarization phenomenon.…”

“…With the intervention of secondary information, under the influence of strong positive information, most individuals start to develop positive attitudes, and only a few individuals maintain negative attitudes, but individual attitudes mostly range from 0 to 0.6, and no positive polarization phenomenon is formed. To sum up, it can be found that, although the model proposed in the literature [ 14 ] simulates the phenomenon of public opinion reversal, it fails to truly reflect the evolution of attitudes of extreme individuals, so it is impossible to describe the phenomenon of reversed public opinion in reality. Contrarily, the model mentioned in this paper not only simulates the phenomenon of public opinion reversal, but also reasonably reflects the distribution and changes of extreme individuals’ attitudes, which is more in line with reality.…”

“…In order to further compare the differences with other models, the public opinion reversal model proposed in the literature [ 14 ] was used to simulate the “Liu Qiangdong sexual assault”, and the advantages of the model were further illustrated by comparing the simulation effects of the two models on cases. The simulation results of the model in the literature [ 14 ] are shown in Figure 33 and Figure 34 .…”