Fitting the complexity of the social space into the right dimension

Complexity within current societies must be addressed by considering relevant both the right dimension as well as relevant factors such as spatial scale, context, and heterogeneity of the data.
Fitting the complexity of the social space into the right dimension
Like

Highlight

There is a need for understanding how current societies are confronted with increasing inconsistencies and contradictions. The recent ability for studying massive datasets has unscrambled complex social dynamics. This complexity within current societies must be addressed by considering relevant both the right dimension as well as relevant factors such as spatial scale, context, and heterogeneity of the data.

Behind the Paper

Today's world is hard to understand. We have more information than ever before and we grapple with even more contradictions. Let's think about the ongoing COVID-19 pandemic and try to find out if relatively speaking the virus impacted mainly the heart of the most populated cities, where people live more concentrated... or those low densely populated areas but with strong social ties. On a political basis, the rise of populist movements in many Western countries over the last decade has been interpreted as the rebellion of the rural world, but how could it be true if the people working in primary activities and living in these countries represent a relatively tiny population segment. Recently, we face an open discussion about global population growth whose approaches range from the Malthusian theories on global overpopulation down to the unsustainability of future pensions in most of the Western countries showing sub-replacement fertility rates for many years. We can move from these great societal dilemmas to others most focused on the individuals. What would be better for a migrant from an economical perspective: to work in a relatively less wealthy city with lower associated living costs or to move to a richer city with very high living costs?

In social sciences, most of the scientific findings rely on experiments from small empirical datasets. In the last decade, the explosion of big data questions many of these findings by framing them into very complex social dynamics [1]. To better address this, social sciences must consider factors such the spatial scale, context, and heterogeneity. Scale refers to the size or length at which a phenomenon or process occurs. Thus, results from geospatial analysis depend on the spatial aggregation unit, which leads to one of the most, longstanding and far-reaching problems in geography: the modifiable areal unit problem, aka MAUP [2]. The relation between the study area and its nearby region determines the context. For instance, one can be rich in a specific neighborhood, but poor in comparison to the overall nation. This depends precisely on context. Finally, heterogeneity shows the diversity of information. For instance, finding poor people in rich nations and rich people in poor nations. In all the cases, the way data are aggregated, in both shape and scale, determines the results.

Recent studies address complexity related to human dynamics. Here we quote several examples. Referring to population's spatial distribution, urbanization processes are not exclusive to the largest and most populated cities, but also emerge in remote and relatively rural regions where small towns act as child-size cities by concentrating most of the nearby population over time [3]. Related to the socioeconomic interactions between societies, Hidalgo [4] argued that the evolution of social and economic systems is a consequence of the physical embodiment of information. Referring to human mobility, studies have found a contradiction between individual and collective mobility patterns [5]. On one hand, many studies argue that human mobility is described as scale-free, whereas others argue that mobility patterns contain meaningful scales making it possible to somewhat predict human behavior [6,7].

Social space refers to physical or virtual places where people interact with one another, which decisively influences collective behaviors. However, many unknowns exist about the nature and complexity of the social space, or its relationship to context and spatial scale. In our recently published paper in Scientific Reports [*] we illustrate the complexity of the social space as a multiscale function using data related to human interactions and income composition. For this, we aggregate fine-grained data collected from both official (US Census Bureau) and unofficial (Twitter) data sources across multiple scales showing substantial changes in the observed patterns. Based on Twitter data, we run an algorithm for community detection ranging from supranational to regional/local clusters. Data from US Census Bureau allow for representing the US income composition ranging from coastal-inland divergence to urban segregation patterns. In all these examples, we show how the spatial patterns are strongly dependent on the scale of the analysis.

Recent studies discuss the risk of oversimplification and binary thinking to understand the complex reality of current societies. Our work agrees with the risks outlined in these studies and raises the need for holistic and systemic approaches in social sciences to better address this complexity. Social scientists must consider the existence of multiscale patterns of information and problems derived from MAUP in mapping, but also the high spatial heterogeneity of human behavior. Therefore, social studies must be framed into the right dimension to avoid biased results and/or misleading conclusions. Biased data analysis may lead to the adoption of fragile and poor decisions. In particular, this concern must be better emphasized with the increasing importance of Artificial Intelligence, as some experts point to the need to develop an Anthropological Intelligence with a sense of social context [8].

Human communicability via Twitter across the United States mainland. From top to bottom: (top) flows of interconnectivity between users via mentions or retweets, and detection of (middle) communities and (bottom) sub-communities.

Income composition in the United States mainland by considering different aggregation levels—from 2 to 1000 km. Regions below national average income are shown in blue, while regions above national average income are shown in red. White-colored regions that emerge as gaps show regions with similar values to the national average.

Authors and affiliations

José Balsa-Barreiro1,2, Mónica Menéndez2 and Alfredo J. Morales1,3

1MIT Media Lab, Massachusetts Institute of Technology, 75 Amherst Street, Cambridge, MA 02139, United States.

2Division of Engineering, New York University Abu Dhabi, P.O. Box 129188, Saadiyat Island, Abu Dhabi, United Arab Emirates.

3New England Complex Systems Institute, 277 Broadway, Cambridge, MA 02143, United States.

Acknowledgements

J. Balsa-Barreiro and M. Menendez acknowledge the support of the NYUAD Center for Interacting Urban Networks (CITIES), funded by Tamkeen under the NYUAD Research Institute Award CG001. J. Balsa-Barreiro and A. Morales acknowledge MIT Media Lab for the support during the time this research was conducted. A. Morales acknowledges the support of the New England Complex Systems Institute (NECSI).

References

  1. Balsa-Barreiro, J., Morales, A.J. & Castelló, E. (2018): Datos, Inteligencia Artificial y complejidad. Una visión de la sociedad del futuro. Invited presentation at the Instituto de la Ingeniería de España, Madrid (Spain): Nov. 05th, 2018. Accessible online at the IIES webpage (In Spanish).
  2. Carballada, A.M. & Balsa-Barreiro, J. (2021): Geospatial analysis and mapping strategies for fine-grained and detailed COVID-19 data with GIS. ISPRS International Journal of Geo-Information 10(9), 602. https://doi.org/10.3390/ijgi10090602
  3. Balsa-Barreiro, J., Morales, A.J. & Lois, R.C. (2021): Mapping population dynamics at local scales using spatial networksComplexity, Article ID 8632086. https://doi.org/10.1155/2021/8632086
  4. Hidalgo, C. (2015): Why information grows? The evolution of order, from atoms to economies. New York, United States: Ed. Basic Books.
  5. Alessandretti, L., Aslak, U. & Lehmann, S. (2020): The scales of human mobility. Nature 587, 402-407.
  6. Song, C., Qu, Z., Blumm, N. & Barabasi, A. (2010): Limits of predictability in human mobility. Science 327, 1018-1021.
  7. Loder, A., Ambühl, L., Menendez, M. & Axhausen, K.W. (2019): Understanding traffic capacity of urban networks. Scientific Reports 9, 16283. https://doi.org/10.1038/s41598-019-51539-5
  8. Dutton, K. (2020): Black-and-white thinking: The burden of a binary brain in a complex world. London: Ed. Picador.

 * Balsa-Barreiro, J., Menéndez, M. & Morales, A.J. (2022): Scale, context, and heterogeneity: the complexity of the social space. Scientific Reports 12, 9037. https://doi.org/10.1038/s41598-022-12871-5