The Impact of Culture on the Individual Subjective Well-Being of the Italian Population: An Exploratory Study

Abstract

The aim of this study is to explore the relationship between cultural access and individual psychological well-being, in order to provide a possible estimation of the impact of cultural participation upon subjective perceptions of well being. Our exploratory research was based on a cross-sectional survey undertook on a medium-large sample (n = 1500) of Italian residents in fall 2008. We refer to the Psychological General Well-Being Index–PGWBI, a tool that has been validated through 30 years of research, as an index of measurement. Moreover, we have administered to the sample an additional questionnaire inquiring about access to 15 distinct culturally related activities. Data are processed by means of a specific methodology based on ANN and Called TWIST. TWIST has been developed by the Semeion Research Center, Rome. Our analysis suggests that culture has a relevant role as a determinant of individual psychological well-being, in that a selected subset of cultural variables turn out to perform among the best predictors of individual PGWB levels. Our results also allow some preliminary considerations about innovative, well-being focused public health policies leveraging upon the human and social developmental role of culture.

Appendices

Appendix 1: The TWIST Methodology

TWIST (Buscema 2005) is an ensemble of two distinct algorithms: T&T and I.S.

T&T

The “Training and Testing” algorithm (T&T) is based on a population of n ANNs, managed by an evolutionary system. In its simplest form, this algorithm reproduces several distribution models of the complete dataset DΓ (one for every ANN of the population) in two subsets ($d_{\Gamma }^{{\left[ {tr} \right]}}$, the Training Set and $d_{\Gamma }^{{\left[ {ts} \right]}}$, the Testing Set). During the learning process each ANN, according to its own data distribution model, is trained on the subsample $d_{\Gamma }^{{\left[ {tr} \right]}}$ and blind-validated on the subsample $d_{\Gamma }^{{\left[ {ts} \right]}}$.

The performance score reached by each ANN in the testing phase represents its “fitness” value (i.e., the individual probability of evolution). The genome of each “ANN-individual” thus codifies a data distribution model with an associated validation strategy. The n data distribution models are combined according to their fitness criteria using an evolutionary algorithm. The fitness-based selection of “ANN-individuals” determines the evolution of the population; that is, the progressive improvement of performance of each network until the optimal performance is reached, which is equivalent to the optimal splitting of the global dataset into subsets. The evolutionary algorithm ruling this process, named “Genetic Doping Algorithm” (GenD), is similar to a genetic algorithm (i.e. it works by crossover and mutation genetic operators), but maintains a constitutional instability across the evolutionary process, thereby sustaining a natural proliferation of biodiversity and a continuous meta-evolution of the population.

The working of T&T is organized into two phases:

• 1) Preliminary phase: In this phase, an evaluation of the parameters of the fitness function that will be used upon the global dataset is performed. During this phase, an inductor ${\Omega_{{D_{\Gamma }^{{\left[ {tr} \right]}},A\,,F\,,Z}}}\left( \cdot \right)$ is set up, which consists of an artificial neural network equipped with a standard Back Propagation algorithm. For this inductor, the optimal configuration is determined at the end of different training trials on the global dataset D_Γ. In this way, the configuration that most “suits” the available dataset is determined: The number of layers and hidden units, and some possible generalizations of the standard learning law. The parameters thus determined define the configuration and the initialization of all the ANN-individuals of the population, and will subsequently stay fixed in the following computational phase. Basically, during this preliminary phase there is a fine-tuning of the inductor that defines the fitness values of the population’s individuals during evolution.

  The accuracy of the ANN performance upon the testing set will be the fitness of that individual (that is, of the trial-specific tentative distribution into two halves of the whole dataset).

• 2) Computational phase: The system extracts from the global dataset the best training and testing sets. During this phase, the ANN-individuals carry out their computational task, based upon the established configuration and the initialization parameters. From the evolution of the population, managed by the GenD algorithm, the best distribution of the global dataset DΓ into two subsets is generated, starting from the initial population of possible solutions $x = \left( {D_{\Gamma }^{{\left[ {tr} \right]}},D_{\Gamma }^{{\left[ {ts} \right]}}} \right)$. Preliminary experimental sessions are performed using several different ANN initializations and configurations, in order to achieve the best partition of the global dataset.

I.S.

Parallel to T&T, TWIST runs I.S. (Input Selection), an adaptive system which is also based on the evolutionary algorithm GenD, and that is able to evaluate the relevance of the different variables of the dataset in a sophisticated way. Therefore, it can be considered as a tool at the same level as a feature selection technique.

From a formal point of view, I.S. is an artificial organism based on the GenD algorithm, and consists of a population of ANNs, in which each one carries out a selection of the independent variables on the available database. The elaboration of I.S., as for T&T, is developed in two phases:

• 1) Preliminary phase: An inductor ${\Omega_{{D_{\Gamma }^{{\left[ {tr} \right]}},A\,,F\,,Z}}}\left( \cdot \right)$ is configured to evaluate the parameters of the fitness function. This inductor is a standard Back-Propagation ANN. The parameters configuration and the initialization of the ANNs are carried out with particular care to avoid possible over-fitting problems that can be present when the database is characterized by a large number of variables that describe a small quantity of data. The number of epochs E₀ necessary to train the inductor is determined through preliminary experimental tests.

• 2) Computational phase: The inductor carries out its computational task, with the configuration determined in the previous phase and the fixed initialization parameters, to extract the most relevant variables of the training and testing subsets. Each ANN-individual of the population is trained on the training set $D_{\Gamma }^{{\prime \left[ {tr} \right]}}$ and tested on the testing set $D_{\Gamma }^{{\prime \left[ {ts} \right]}}$.

The evolution of ANN-individuals in the population is again based on GenD. In the I.S. approach, the GenD genome consists of n binary values, where n is the cardinality of the original input space. Every gene indicates whether the corresponding input variable is active or not in that particular selection of variables. For each genome, the relevant fitness value is computed as usual. Through the evolutionary algorithm, the different “hypotheses” of variable selection, generated by each ANNs within the population, change over time, at each generation: This leads to the selection of the best combination of input variables. As in T&T, the crossover and mutation genetic operators are applied upon the ANNs population; the rates of occurrence for both operators are adaptively self-determined by the system at each generation.

When the evolutionary algorithm no longer improves its performance, the process stops, and the best selection of the input variables is employed on the testing subset. In order to improve the speed and the quality of the solutions that have to be optimized with respect to standard evolutionary algorithms, GenD does not breed the best-performing ANN-individuals, but rather the most representative ones. The selection criterion is therefore not that of picking up momentarily brilliant but possibly unreliable outliers, but rather reinforcing those characteristics that are stably well performing.

Appendix 2. List of Diseases Included in the Questionnaire

Hypertension

Heart Attack

Heart Diseases

Diabetes

Angina

Cancer

Allergy

Arthritis

Low Back Pain

Lung Diseases

Skin Diseases

Deafness

Limited Arms and/or Legs Functionality

Blindness

Psychiatric Disturbances

Depression