The object of the first study is to assess the ability of people to estimate their confidence in answering binary questions, and the dependence of these estimates on the methods of their measurement. The main goal was to examine the nature of excessive confidence of the individual, including a novel approach given by duplex comparison mechanisms.3. Methods of Research
We set up an experiment that compares the calibration of predictions using a standard scale and duplex scale; we contrast the effects of different behavioral methods of calibration measurement – namely, though self-reporting and through betting on the correct answer. Furthermore, we compare and teste the validity of different theoretical neuroeconomic models of decision - Diffuse vs. Race models of decision-making. Finally, we compare calibration forecasts using two samples: high-school students and university students.4. Results of Research
The study identified a number of results, the most significant of which concern the method of measuring the relationship between calibration and confidence, and the ensuing level of calibration. Specifically, in common single-scale calibration tasks the level of calibration uniformly decreases with the confidence of the respondent, while with the duplex scale this dependence is parabolic: at a low confidence level, an increase in confidence is positively related to the quality of forecast, while at high confidence level this relationship is negative. Furthermore, bidding rates were more reliable predictors of the quality of predictions than verbal scores, and coincide with the latter only when the measures of the announced confidence under duplex scale in both cases were consistent.
In the second study we explore the possibilities of applying structural models of individual behaviour to survey experiments in a real world environment – to model transportation behaviour of the citizens of Moscow. Determining an optimal strategy of development of the city's transport system is an important issue of urban planning, especially in a large city with high congestion, such as Moscow. It is important to learn the motives of travelling citizens in order to understand how the city should optimally manage the traffic flows under physical constraints of limited road space, and opportunities for the development of in-city rapid transport system (street and freeway networks, parking space, the possibility of transport regulations, etc ) . In current literature there are two main approaches to this problem . The first of them - physical - considers public transport as a dynamic system in which the interaction of physical particles (atoms ), which act as individual cars. Such studies use dynamic models for the study of particles, such as the phase flows of cars in a given place at a given time of day, hydrological models to determine their redirection in case of congestion, stochastic processes and random graphs to predict the probability of their occurrence in a given part of the network, etc. These models often concur with the equilibrium representations ( in the sense of Nash ) of behavior of typical (representative ) drivers, assuming other drivers make their decisions based on the same considerations. At the same time, these models, do not take into account the fact that the particles in this case are endowed with free will to accept or reject a particular strategy under particular circumstances. Yet more importantly, individuals can change their behavior depending on the sequences and contents of policy interventions, such as improvement of public transportation systems or changes in city moods and fashions. How to change these preferences through economic or legal restrictions on the use of the road network? What would be the residents' attitudes to these alternative policies ? How will the transport behavior depend on the weather conditions, level of congestion, quality of public transport, available information, or alternative pricing models for transportation services? These problems can be addressed by structural models of demand for transport infrastructure, which evaluate this demand taking into account the characteristics of the consumers.5. Field of usage
In a first study of its kind in Moscow, we apply these models to study the behavior of the city's inhabitants, and then conduct a pilot survey ( field experiment) in order to identify their preferences and attitudes to different scenarios of urban transport policy. Using this pilot, we were able to identify a number of factors that seem to have an impact on the city residents' decisions to choose private vs public transport, as well as to evaluate the effects of some policies implemented by the city authorities. Specifically, we found that people tend to prefer private rather than public transport if they reside in a remote place of the city (far away from the metro station), if the road network is characterized by low connectivity, if time list in traffic jams is high, and if they view public transport as uncomfortablee. Estimates of the elasticity of demand for public transport with respect to these factors allow us to predict the expected response to the residents on the specific transport policy measures .