I would like to develop an agent-based model in which agents are imposed with cognitive constraints.
I suppose to initialize these constraints by means of a special distribution that tells us how many individuals can handle, say, 5 posts on social media per day, how many individuals can capture 10 posts per day, and so on.
Are there any empirically supported distributions that can model this issue?
It seems like that the majority of people can handle only limited amount of information (recall the Dunbar number), but at the same time, I suppose, there are individuals who can (and who are willing to) proceed much more content. For example, many of us read posts and only top comments (suggested by ranking algorithms) to them on social media, whereas others read all comments.
The question is whether such patterns have been already analyzed in quantitative empirical studies. In particular, it would be interesting to figure out if individuals who read many posts on social media make the objective distribution heavy-tailed or not.
Dear Ivan,
I have not ever met the research of that kind. I am not an Information Science scholar, though :)
However, what do you mean when saying proceed and handle social media content?
- Just scrolling the feed;
- Scroll and set the eyes on smth;
- Scroll, set the eyes on, and read;
- Read and comprehend?
As for cognitive constraints, I would suggest to consider attention span and working memory capacity. Perhaps, eye tracking measures will work well here.
In theory, behavioral issues such as FOMO (fear of missing out) or social media addiction could be also considered as those which impacted the proposed distribution.
Anastasia Peshkovskaya, many thanks for your answers. Yes, I agree that the current formulation needs clarification. I meant the process of reading.
Though, there is one important facet of this process in the case of social media. When you read, say, a post on social media, you can see only a few comments to it (usually, the most popular). To see the others, you need to push the corresponding button.
And the additional question is what fraction of individuals does push this button and how this depends on users' engagement with the topic of discussion or other covariates.
If "handle" means somre kind of reaction, say "Reply", then, I guess, you willdo your own empirical study
If "handle" is "attend", then it looks hopeless. Though, may be you'll find some data about the distribution of time people stick daily to social media and use it as a proxy for the intended ditribution.
Actually, I am not sure the limitation is really "cognitive". May be, they just cannot spend more time hanging out on social media.
Alexander Petrov , if handle means behavioral reaction like commenting post or smth else, Ivan can surely conduct a social media metrics-based study. If handle means not only read/ comment/ press like or dislike but pay attention, empirical study may either be conducted with eye tracking technology. Eye gaze behavior and its metrics can be used to model various patterns of user behavior on social media.
Sometimes people react with many different emotions and thoughts, but never show that they were triggered by the content.
Ivan Kozitsin thanks. Your question (below) when resolved worth to be published in Science!
And the additional question is what fraction of individuals does push this button and how this depends on users' engagement with the topic of discussion or other covariates.
Alexander Petrov Many thanks for your answer! Yes, I agree that the word "cognitive" is way inaccurate.
No doubt, many factors govern such processes, including to what extent an individual is engaged with the topic of discussion.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India