Archive for the ‘Experiment’ Category


Randomization and its constraints: A critical look at the current reserach practices in social psychology

August 29, 2014

Taku Iwatsuki

Abstract: In this talk, I investigate the importance of randomization in the context of social psychological research. I compare the costs and the benefits of randomization, and argue that the current research practices in social psychology seem to put too much emphasis on the use of randomization. The talk will be structured as follows: First, I briefly explain what randomization is and what role it typically plays in social psychological research. Next, I explore its methodological benefits in the context of causal inference through critical examination of arguments for randomization. Then, I investigate what the costs of randomization are, focusing on the constraints it imposes on other aspects of research design and eventually on psychological theorizing based on such research. Finally, comparing these costs and benefits, I conclude that the use of more diverse research designs with less emphasis on the use of randomization seems to be necessary for developing better psychological theories.



Equivalence Principle Tests

April 3, 2014

Nora Mills Boyd

Abstract: The Equivalence Principle supposedly plays a central role in characterizing the theory of general relativity and in particular, the geometric interpretation of gravity.  Moreover, null results from decades of experimental research looking for violations of the EP contribute support to the claim that GR has passed all experimental tests so far.  However, there is significant disconnect between formulations of the principle in theoretical physics and philosophical literatures on one hand, and experimental practice on the other.  Precise formulations of the principle typically apply strictly to abstract or highly idealized systems involving force-free ‘test’ bodies, perfectly homogeneous gravitational fields, and infinitesimally small regions. In striking contrast, EP experimentalists measure and manage forces, account for tidal effects, and capitalize on (or compensate for) the inhomogeneous gravitational environments of real, physically extended, laboratories.  So what do null results from experiments searching for EP violations actually tell us about GR?  The present project aims to bridge the gap between the conceptual foundations of GR and the experiments that supposedly support that theory.  To do this, I emphasize the importance of measuring and engineering particular gravitational environments in EP experiments.


The Multiplicity of Protocols

February 28, 2014

David Colaco

Abstract: I investigate what implications (1) effects being dependent on experimental protocols, and (2) a multiplicity of experimental protocols, have for convergence and generalization – which I refer to jointly as extension – of findings in neuroscience and other sciences. Together, these claims raise concerns over whether we can compare findings across laboratories, or generalize them to the world. I introduce two successful cases of extension despite different protocols. These cases elucidate what kinds of differences are relevant to extension. In particular, I focus on the role the research purpose plays in understanding which differences make a difference and which do not.