MinireviewExtinction learning, reconsolidation and the internal reinforcement hypothesis
Section snippets
Reconsolidation—a phenomenon
After memory retrieval, an initially consolidated memory becomes susceptible to amnestic treatments, like the application of an electric shock, treatment with protein synthesis blockers, blockers of metabolic processes or amnestic agents. This results in a decrease of the conditioned response at a later retrieval trial (Eisenberg et al., 2003, Litvin and Anokhin, 2000, Misanin et al., 1968, Nader et al., 2000, Pedreira et al., 2002, Przybyslawski and Sara, 1997). Since the retrieval trial
Reconsolidation—the concepts
Two basic concepts have been put forward so far to explain the reconsolidation phenomena: retrieval and the storage deficit.
Several studies found that the reconsolidation phenomenon is only temporary. The behavioral effect reverses with time or by the application of the US (unconditioned stimulus) which some authors interpret as a temporarily blocked memory retrieval rather then a storage deficit (Anokhin et al., 2002, Eisenberg and Dudai, 2004, Lattal and Abel, 2004, Litvin and Anokhin, 2000,
Memory retrieval induces two opposing processes
A second process induced by exposure to the learned stimulus without reinforcement is extinction. The behavioral phenomenon of extinction is defined as a decrease in the conditioned response (CR) resulting from repeated CS presentations (Pavlov, 1927). This decrease of the CR is thought to be based on new learning (“the CS is not followed by the US”) that leads to extinction memory via a protein synthesis-dependent consolidation process (Berman and Dudai, 2001, Bouton and Moody, 2004, Pedreira
The internal reinforcement hypothesis
What might be the mechanisms behind these two memory traces? One trace certainly represents an extinction memory. The other—as expressed in spontaneous recovery from extinction—is less clear. Here, we offer an interpretation based on the properties of the reward system activated by appetitive stimuli. Our interpretation and the resulting internal reinforcement hypothesis are based on the assumption that the phenomenon of retrieval-induced amnesia is due to an inhibited consolidation process.
Interaction of reminder and extinction memory
The internal reinforcement hypothesis could explain why in the honeybee the presentation of multiple retrieval trials leads to spontaneous recovery of extinction resembling the reconsolidation phenomenon. Following our proposal that reminder learning resembles second-order conditioning with the CS being CS1 and CS2 at the same time, reminder learning should have the same properties as second-order conditioning. Second-order conditioning is transient. It reaches an asymptote after a small number
The internal reinforcement hypothesis and the original memory
The traditional Pavlovian view of spontaneous recovery envisages that some of the initial CS-US memory survives the extinction procedure and is temporarily suppressed by the extinction learning (Pavlov, 1927). If the internal reinforcement hypothesis holds true, it is not a temporarily suppressed original CS-US memory, but rather the newly consolidated reminder memory that resembles the spontaneous recovery phenomenon. But then what happened to the original CS-US memory? If the newly
Is the honeybee different?
In contrast to the honeybee study, several studies demonstrate an induction of the reconsolidation phenomenon after one retrieval trial and the extinction of a CR after many retrieval trials (see above). If spontaneous recovery resembles the reconsolidation phenomenon, how can this inconsistency be reconciled with the internal reinforcement hypothesis? Spontaneous recovery depends on the number of retrieval trials in a non-linear fashion (Rescorla, 2004). After minimal extinction spontaneous
Concluding remarks
Reconsolidation and extinction phenomena are conserved phenomena that have been found in several vertebrate and invertebrate species. Hence, it can be concluded that these phenomena resemble basic properties of memory formation. It is still unknown which neural mechanisms underlie these properties. Brain architecture differs greatly between vertebrates and invertebrates and varies considerably between different invertebrate species (e.g. honeybee and snail, Sangha, Scheibenstock, & Lukowiak,
Note added in proof
Meanwhile Duvarci, Mamou, and Nader (2006) demonstrated in vertebrates that the reconsolidation phenomenon occurs although extinction took place. This underlines our conclusion that extinction is not a boundary condition for reconsolidation.
Acknowledgments
We thank Nicola Stollhoff for fruitful discussions and her contribution to the internal reinforcement hypothesis, our anonymous referee for inspiring criticism, and Mary Wurm for help with this manuscript. This work was supported by the Volkswagen Stiftung (I/77378) and the DFG (EI 512/1-2) with funding for D.E.
References (48)
Mechanisms of memory stabilization: are consolidation and reconsolidation similar or distinct processes?
Trends Neuroscience
(2005)- et al.
Memory processes in classical conditioning
Neuroscience and Biobehavioural Reviews
(2004) - et al.
Cellular and systems reconsolidation in the hippocampus
Neuron
(2002) Reconsolidation: the advantage of being refocused
Currurent Opinion in Neurobiology
(2006)- et al.
Rites of passage of the engram: reconsolidation and the lingering consolidation hypothesis
Neuron
(2004) - et al.
Effects of protein synthesis inhibition on memory for active avoidance training
Physiology & Behavior
(1975) - et al.
The influence of duration of protein synthesis inhibition on memory
Physiology & Behavior
(1973) The neural basis of associative reward learning in honeybees
Trends Neuroscience
(1997)- et al.
Cognitive architecture of a mini-brain: the honeybee
Trends in Cognitive Sciences
(2001) - et al.
Temporally graded requirement for protein synthesis following memory reactivation
Neuron
(2002)