---
bibtex: @article{Rosenwein1995-ROSHHA,
title = {Heuristics, Hypotheses, and Social Influence: A New Approach to the Experimental Simulation of Social Epistemology},
author = {Robert Rosenwein and Michael Gorman},
journal = {Social Epistemology},
publisher = {Taylor \& Francis},
pages = {57--69},
volume = {9},
number = {1},
doi = {10.1080/02691729508578774},
year = {1995}
}
---
Heuristics, hypotheses, and social influence: a new approach to the experimental simulation of social epistemology
This is an analogue simulation of science using students, not a computer simulation.
This paper is one of two exploring a strategy of experimental simulation of science-like knowledge generation that it is believed has greater realism or ecological validity than present laboratory simulations. The simulation is called SIMSCI or 'Simulated Science', loosely based on SIMSOC or 'Simulated Society', developed by William Gamson and his associates in the late 1960s and used extensively since then in both research and educational contexts. In the present paper interesting and at the same time disappointing results of a study are presented. It was this disappointment, discussed at the end of the paper, which stimulated consideration of the development of SIMSCI. In the second paper, SIMSCI is discussed in more detail, suggesting how it might be a platform that allows researchers to explore the relationship of different kinds of social arrangements, networks of alliances, the operation of social influence and negotiation, and mental representations as they contribute to knowledge production.
Aim is to present a hybrid methodology (vivo & in vitro) using analog simulation to test the boundaries of normative assumptions about science, and shed light on the debate about the role of the natural world in determining the confidence of scientists in their knowledge claims.
The thrust of our efforts is to make it both possible and convincing to use insights about the relation of social/cognitive factors and scientific decision-making gener- ated in in vitro studies to illuminate in vivo studies—and vice versa. (p57)
Implicationsfor simulatingscience ...
a start has been made down the path of trying to increase the ecological validity of the simulation and the attempt is disppointing. (p67)
he task used here is science-like in that it asks subjects to generate falsifiable data and by analogy mimics the activities of scientists in proposing hypo- theses and generating tests of these hypotheses. On the other hand, 'science-like' is not the same as 'science'. Would not experimental realism be better served if ways could be found of translating actual scientific problems into forms that could be worked with in a more controlled setting?
The virtue of simplification is the ability to make stronger causal inferences; indeed, what gives science as it is practiced its raisond'être but this very property?
The problem is that between experimental research at one end and case studies at the other there does not exist that middle (perhaps too optimistic) ground that balances controlled environments with the case study itself. It is believed that it may be possible to develop such a middle ground; a start has been made here. It is called simulatedscience and the companion paper to this one sketches out preliminary thoughts about how it might work.