Related Experiment Videos
A Turing test of a timing theory.
Russell M Church1, Paulo Guilhardi
1Department of Psychology, Box 1853, Brown University, Providence, RI 02912, USA. Russell_Church@Brown.edu
Behavioural Processes
|March 30, 2005
Summary
This study introduces a Turing test to evaluate quantitative theories of timing and conditioning. Behavioral data from rat experiments reliably matched theoretical predictions, suggesting the test
Area of Science:
- Behavioral science
- Computational neuroscience
- Psychology
Background:
- Quantitative theories of timing and conditioning require rigorous evaluation methods.
- The Turing test offers a novel approach to assess the predictive accuracy of such theories.
Purpose of the Study:
- To evaluate a quantitative theory of timing using a Turing test framework.
- To compare experimental behavioral data with predictions from a packet theory of timing.
Main Methods:
- Rats were trained on fixed-interval reinforcement schedules.
- A simulation of a packet theory of timing was used to generate predictions.
- An objective classification rule assessed similarity between empirical data and theoretical predictions.
Main Results:
- The observed probability of correct classification was 0.6, exceeding the chance level of 0.5.
- This result was statistically significant, indicating reliable differentiation between data and theory.
- The packet theory's predictions showed a slight but consistent advantage.
Conclusions:
- A Turing test provides a graded and reliable metric for evaluating quantitative theories.
- This methodology can be applied to assess the validity of timing and conditioning models.
- The findings support the utility of the Turing test in computational and behavioral sciences.