Abstract
I discuss top-down modulation of perception in terms of a variable Bayesian learning rate, revealing a wide range of prior hierarchical expectations that can modulate perception. I then switch to the prediction error minimization framework and seek to conceive cognitive penetration specifically as prediction error minimization deviations from a variable Bayesian learning rate. This approach retains cognitive penetration as a category somewhat distinct from other top-down effects, and carves a reasonable route between penetrability and impenetrability. It prevents rampant, relativistic cognitive penetration of perception and yet is consistent with the continuity of cognition and perception.
| Original language | English |
|---|---|
| Pages (from-to) | 75-85 |
| Number of pages | 11 |
| Journal | Consciousness and Cognition |
| Volume | 47 |
| DOIs | |
| Publication status | Published - 1 Jan 2017 |
Keywords
- Active inference
- Attention
- Bayesian inference
- Cognitive penetrability
- Hierarchical Gaussian filter
- Illusion
- Prediction error minimization
- Variable learning rate
Cite this
}
Priors in perception : Top-down modulation, Bayesian perceptual learning rate, and prediction error minimization. / Hohwy, Jakob.
In: Consciousness and Cognition, Vol. 47, 01.01.2017, p. 75-85.Research output: Contribution to journal › Article › Research › peer-review
TY - JOUR
T1 - Priors in perception
T2 - Top-down modulation, Bayesian perceptual learning rate, and prediction error minimization
AU - Hohwy, Jakob
PY - 2017/1/1
Y1 - 2017/1/1
N2 - I discuss top-down modulation of perception in terms of a variable Bayesian learning rate, revealing a wide range of prior hierarchical expectations that can modulate perception. I then switch to the prediction error minimization framework and seek to conceive cognitive penetration specifically as prediction error minimization deviations from a variable Bayesian learning rate. This approach retains cognitive penetration as a category somewhat distinct from other top-down effects, and carves a reasonable route between penetrability and impenetrability. It prevents rampant, relativistic cognitive penetration of perception and yet is consistent with the continuity of cognition and perception.
AB - I discuss top-down modulation of perception in terms of a variable Bayesian learning rate, revealing a wide range of prior hierarchical expectations that can modulate perception. I then switch to the prediction error minimization framework and seek to conceive cognitive penetration specifically as prediction error minimization deviations from a variable Bayesian learning rate. This approach retains cognitive penetration as a category somewhat distinct from other top-down effects, and carves a reasonable route between penetrability and impenetrability. It prevents rampant, relativistic cognitive penetration of perception and yet is consistent with the continuity of cognition and perception.
KW - Active inference
KW - Attention
KW - Bayesian inference
KW - Cognitive penetrability
KW - Hierarchical Gaussian filter
KW - Illusion
KW - Prediction error minimization
KW - Variable learning rate
UR - http://www.scopus.com/inward/record.url?scp=84994749945&partnerID=8YFLogxK
U2 - 10.1016/j.concog.2016.09.004
DO - 10.1016/j.concog.2016.09.004
M3 - Article
VL - 47
SP - 75
EP - 85
JO - Consciousness and Cognition
JF - Consciousness and Cognition
SN - 1053-8100
ER -