Related Experiment Videos
Bayesian inference and "truth": a comment on Hoffman, Singh, and Prakash.
1Department of Psychology, Center for Cognitive Science, Rutgers University - New Brunswick, 152 Frelinghuysen Rd, Piscataway Township, NJ, 08854, USA. jacob@ruccs.rutgers.edu.
Psychonomic Bulletin & Review
|September 20, 2015
Summary
This study argues that visual perception does not require or achieve literal truth. Bayesian inference frameworks support this view, suggesting perceptual inferences operate without needing to be factually correct.
Area of Science:
- Cognitive Science
- Neuroscience
- Philosophy of Mind
Background:
- The traditional view posits that the visual system's primary function is to achieve veridicality, or literal truth, in perception.
- Hoffman, Singh, and Prakash challenge this by proposing a framework where perceptual inferences do not rely on factual accuracy.
Purpose of the Study:
- To support the argument that veridicality is not essential for the visual system.
- To propose that Bayesian inference models already align with this non-veridical epistemological stance.
Main Methods:
- Conceptual analysis of existing theories on visual perception.
- Examination of the epistemological implications of Bayesian inference in perception.
Main Results:
- Agreement with the proposition that the visual system does not require or achieve veridicality.
- Identification of Bayesian inference as a framework that inherently supports a non-veridical approach to perceptual inferences.
Conclusions:
- The visual system's inferences may not need to be literally true to be functional.
- Bayesian inference provides a compatible theoretical basis for understanding perception without strict adherence to veridicality.