Seminario "The wisdom of converging and diverging crowds"
Miércoles 16/10, 17.15h
Presentado por Joaquín Navajas
Paper Abstract:
The "wisdom of crowds" refers to the phenomenon where the aggregation of independent judgments often leads to more accurate estimates than individual assessments. This effect has been applied to fields such as medical diagnosis and financial forecasting. However, two contrasting forces can shape the accuracy of collective judgments: social convergence and opinion divergence. In this talk, I will present findings from two complementary approaches to collective decision-making. The first explores how structured deliberation within small, independent groups can enhance collective accuracy. Through a live experiment with over 5,000 participants, we found that group consensus after discussion significantly outperforms the aggregation of initial, independent estimates. Even a small number of consensus decisions outshines the wisdom of large crowds. The second approach challenges traditional thinking by introducing purposeful divergence. Here, anchoring different halves of the crowd to contrasting extremes leads to increased diversity of opinion, ultimately reducing collective error. Across several experiments, this method proved to improve the accuracy of group estimates in estimation and forecasting tasks. By integrating insights from both converging and diverging crowds, this talk will offer a richer understanding of how to harness collective intelligence in various decision-making contexts.
The "wisdom of crowds" refers to the phenomenon where the aggregation of independent judgments often leads to more accurate estimates than individual assessments. This effect has been applied to fields such as medical diagnosis and financial forecasting. However, two contrasting forces can shape the accuracy of collective judgments: social convergence and opinion divergence. In this talk, I will present findings from two complementary approaches to collective decision-making. The first explores how structured deliberation within small, independent groups can enhance collective accuracy. Through a live experiment with over 5,000 participants, we found that group consensus after discussion significantly outperforms the aggregation of initial, independent estimates. Even a small number of consensus decisions outshines the wisdom of large crowds. The second approach challenges traditional thinking by introducing purposeful divergence. Here, anchoring different halves of the crowd to contrasting extremes leads to increased diversity of opinion, ultimately reducing collective error. Across several experiments, this method proved to improve the accuracy of group estimates in estimation and forecasting tasks. By integrating insights from both converging and diverging crowds, this talk will offer a richer understanding of how to harness collective intelligence in various decision-making contexts.
Joaquín Navajas
Joaquín Navajas es profesor-investigador en la Escuela de Negocios de la Universidad Torcuato Di Tella e investigador del CONICET. Es Licenciado en Ciencias Físicas de la Universidad de Buenos Aires y PhD en Neurociencia de la University of Leicester del Reino Unido. Entre 2014 y 2017 fue investigador postdoctoral en el Institute of Cognitive Neuroscience de University College London, universidad número uno en el mundo en neurociencias.
Sus líneas de investigación se centran en entender los sesgos en la toma de decisiones, el comportamiento de equipos de trabajo, las decisiones morales y los procesos cognitivos vinculados a la polarización política. Ha trabajado previamente en diversas líneas relacionadas con la neuroingeniería, la neurociencia cognitiva, la psicología experimental y las ciencias del comportamiento humano.
Sus trabajos de investigación han sido publicados en Nature Human Behaviour, Journal of Neuroscience, Journal of Experimental Psychology y Current Opinion in Behavioral Sciences, entre otros. Ha participado también como orador y consultor en numerosos congresos académicos, coloquios y eventos empresariales. En 2012 fue galardonado con el premio ICT Pioneers, que reconoce a los candidatos doctorales más destacados del Reino Unido. En el ámbito local, fue destacado por La Nación en 2014 como uno de los 35 referentes sub-35 de la Argentina.
Actualmente dirige el Laboratorio de Neurociencia de la Universidad Torcuato Di Tella y es profesor de grado y posgrado en la Escuela de Negocios, además dicta programas sobre neurociencia en el área de Educación Ejecutiva, y tiene a cargo la materia “Neurociencia y Psicología Experimental”, optativa para todas las carreras de grado de la UTDT.