The neural basis of partial reinforcement
-
1
The University of Queensland, Queensland Brain Institute, Australia
The partial reinforcement extinction effect (PREE) is a seemingly paradoxical phenomenon in human and animal learning in which the omission of reinforcement during the acquisition of conditioning produces an increase in conditioned responding during extinction. This phenomenon suggests that in human learning, partial reinforcement would result in more efficient learning and more stable learning. The biology that underpins partial reinforecement is poorly understood. I will describe experiments in an animal model of PREE in which we have tested the effect of partial reinforcement (100% and 50% reinforcement) on fear learning using Pavlovian Fear Conditioning. We have tested the impact of partial reinforcement on learning, and then used cell biology and electroohysiolgoical recordings to understand the cellular changes that underpin PREE. Immunohistochemical analysis revealed increased labelling of p-ERK and c-fos in central and basolateral amygdala, hippocampus, medial prefrontal cortex, auditory cortex and somatosensory cortex that correlated with freezing. Using in-vivo multi-unit microelectrode recording revealed that the size of the PREE depended on oscillatory activity in amygdala and subiculum, particularly in the theta/alpha band (8-10Hz). The magnitude of the PREE was also related to the synchronization of LFP oscillatory activity between BLA and subiculum.
In human subjects neural activity was measured using 64 channel EEG. Behavioural responses to an explicit task were made with keyboard presses. Heart rate, respiration rate and skin conductance responses (SCR) were measured continually throughout the experiment to index both conditioned and unconditioned responses. A within-subject design was implemented, with acquisition and extinction occurring in a single session. The conditions included 100%, random 50% and alternating (i.e. non-random) 50% reinforcement. Crucially, the random 50% condition showed greater resistance to extinction than the 100% or alternating 50% conditions. Increased conditioned responding (SCR and heart rate increases) in the 50% condition was associated with increased power in the EEG theta band (5-10Hz). The results of these experiments revealed several novel findings: (1) A PREE was observed in both Wistar rats and humans using similar auditory CS-shock US fear conditioning protocols. (2) The PREE was demonstrated to be context sensitive: it could effectively be abolished by extinguishing in a different context to acquisition; (3) The PREE correlated with neural activity in amygdala, hippocampus, medial prefrontal cortex, auditory cortex and somatosensory cortex; (4) The magnitude of the PREE correlated with synchronized oscillatory activity between amygdala and hippocampus (subiculum). These results suggest that partial reinforcement learning results within a Bayesean predictive coding model of brain function.
Keywords:
Learning,
animal model,
Cell Biology,
electroohysiolgoical recordings,
p-ERK,
c-fos,
Amygdala,
Hippocampus,
Medial prefrontal cortex,
Auditory Cortex,
Somatosensory Cortex,
in-vivo multi-unit microelectrode recording,
partial reinforcement extinction effect (PREE)
Conference:
International Conference - Educational Neuroscience, Abu Dhabi, United Arab Emirates, 28 Feb - 29 Feb, 2016.
Presentation Type:
Oral Presentation (invited speakers only)
Topic:
Educational Neuroscience
Citation:
Sah
P and
Morris
J
(2016). The neural basis of partial reinforcement.
Front. Neurosci.
Conference Abstract:
International Conference - Educational Neuroscience.
doi: 10.3389/conf.fnins.2016.92.00019
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
25 Jan 2016;
Published Online:
23 Mar 2016.
*
Correspondence:
Prof. Pankaj Sah, The University of Queensland, Queensland Brain Institute, St Lucia, Queensland, 4072, Australia, pankaj.sah@uq.edu.au
Dr. John Morris, The University of Queensland, Queensland Brain Institute, St Lucia, Queensland, 4072, Australia, r.harvey1@uq.edu.au