What is extinction in classical conditioning?

VI. Summary

Operant extinction is a procedure in which reinforcement of a previously reinforced behavior is discontinued. Published studies have demonstrated that the procedure, which is conceptually sound and logically simple, is effective in reducing a wide range of troublesome operant behaviors in a variety of clients. Although extinction is a useful and generally accepted response-reduction procedure, it can be difficult to arrange consistently. Moreover, troublesome behavior may persist for a substantial time, and even briefly increase in rate and intensity, under extinction. These and other considerations occasionally preclude the use of extinction to deal with particular behavioral problems.

Resistance to Extinction Associated with Drug Self-administration

Extinction procedures can provide measures of the motivational properties of drugs by assessing the persistence of drug-seeking behavior in the absence of response-contingent drug availability. In an extinction paradigm, subjects are trained to self-administer a drug until stable self-administration patterns are achieved, and then the drug is removed (Schuster and Woods, 1968) (Fig. 2.16). Extinction testing sessions are identical to training sessions except that no drug is delivered after completion of the response requirement. Measures provided by an extinction paradigm reflect the degree of resistance to extinction and include the duration of extinction responding, the total number of responses emitted during the entire extinction session, and the probability of reinitiating responding under extinction conditions at a later time after successful extinction of the self-administration behavior (i.e., propensity to relapse).

Conclusion

Extinction learning and the molecular mechanisms underlying extinction memory formation are surprisingly similar between honeybees and vertebrates. In honeybees and in vertebrates, a context dependency of extinction has been shown. Furthermore, a reactivation and reconsolidation of the initially formed acquisition memory and the existence of two contrasting memories after extinction have been demonstrated. Recent results on extinction in honeybees extend these findings by showing that extinction memory formation (i.e., its underlying molecular mechanisms) is dependent on the learning parameters of reward learning. This might ensure that only meaningful changes of the reinforcing stimulus are memorized.

Extinction resembles a basic learning phenomenon that enables animals to adequately react to a fluctuating environment by providing them with the ability to learn about environmental changes and memorize this information. The similarity of extinction learning and its underlying molecular mechanisms between honeybees and vertebrates suggests that extinction is a conserved, phylogenetically old mechanism. Given such conserved functions, findings on the mechanisms of extinction in honeybees will be most important to further elucidate the basic mechanisms of extinction learning in both invertebrates and vertebrates.

Introduction

Extinction occurs when a conditioned stimulus (CS) is repeatedly presented in the absence of reinforcement; it is measured as a decline in the frequency and amplitude of conditioned responses. Current theories of extinction suggest that this decline reflects the development of an inhibitory memory that competes with the conditioning memory to drive behavior. Like other forms of memory, extinction consists of three phases: acquisition, consolidation, and retrieval (Figure 1). Acquisition of extinction is evident as decreased responding to the CS across an extinction session. Consolidation begins some time during the extinction session and continues for several hours afterward to form a more stable, long-term memory trace. Retrieval of extinction is triggered when the CS is once again presented after the extinction session. Research on the mechanisms of extinction has pinpointed many of the molecular mechanisms involved in these processes and has identified a critical neural circuit, which includes the amygdala, medial prefrontal cortex (mPFC), and hippocampus.

Extinction recall

Extinction recall examines the between-session strength and persistence of the extinguished association between the CS +-US (i.e., the extinction memory) formed during extinction learning. Extinction recall does not involve new learning, but rather the ability of an individual to retrieve a previously learned extinction memory. In a laboratory setting, extinction recall with human participants typically takes place 24 h after fear extinction and involves the presentation of the CS + without air blast reinforcement (US). This process determines whether a conditioned fear response has remained extinguished or has returned (see “Return of fear” in the section). Psychophysiological research in previously traumatized populations with PTSD has shown deficient between-session extinction recall or a reduced ability to retrieve the extinction memory while suppressing the original fear memory (CS-US association; Acheson et al., 2015; Milad et al., 2009; Shvil et al., 2014). When extinction recall is unimpaired, it is comparable with symptom reduction that is maintained after treatment interventions, most commonly prolonged exposure therapy. Conversely, when extinction recall is impaired, it serves as a model for treatment resistance or relapse, as symptom reduction and clinical gains do not persist posttreatment (Zuj & Norrholm, 2019).

Transcription

Extinction has been associated with an increase in the expression of immediate-early genes in mPFC and BLA and an upregulation of gephyrin and BDNF mRNA in the BLA. A persistent impairment of extinction of step-down avoidance has been found when the transcription inhibitors a-amanitin or 5,6-dichloro-1-b-D-ribofuranosyl benzimidazole (DRB) were infused into the CA1 region of hippocampus 15 min before, but not 1 or 3 h after, the first of several extinction exposures to the apparatus. The lack of effect of post-extinction training administration is somewhat surprising given the likelihood that mRNA transcription is involved in consolidation of extinction memory, although it is possible that the onset of transcription occurs fairly rapidly (<1 h) following extinction training. Facilitation of extinction of fear-potentiated startle by pre-extinction training DCS was blocked by pre-administration of the transcription inhibitor actinomycin D into the BLA, an effect not due to state dependency.

Extinction is the decrease in strength of a learned behavior when the conditioned stimulus is presented without the unconditioned stimulus (in Pavlovian learning), or when the behavior is no longer reinforced (in operant or instrumental learning). This chapter reviews the behavioral literature on extinction. The first section reviews several relapse phenomena, such as the renewal effect, which suggest that extinction does not destroy original learning, but instead involves new context-dependent learning. The second section reviews theoretical explanations of the cause of extinction. The final section reviews behavioral and pharmacological ways of optimizing extinction learning and decreasing the likelihood of relapse.

Conclusions

Extinction is a common neurologic deficit that often occurs as one of a constellation of symptoms seen with lesions of the PPC. Although extinction has typically been considered a deficit in the allocation of attention, new findings, particularly from NHP studies, point to one potential and important source of extinction as damage to decision-making circuits for actions within the PPC. This new understanding provides clues to potential therapies for extinction. Also the finding that the PPC is important for action decisions and action planning has led to new neuroprosthetic applications using PPC recordings as control signals to assist paralyzed patients. Damage to the action apparatus in PPC causes extinction; on the other hand, decoding from a functioning PPC can provide the action signals for controlling BMIs.

Extinction

Extinction (EXT) occurs when the contingency between a target behavior and its reinforcing consequence is interrupted. EXT involves withholding the consequent event maintaining the problem behavior upon its occurrence or making the consequent noncontingent on the target behavior (Granpeesheh, Tarbox, & Dixon, 2009; Miltenberger et al., 2016). EXT may take several forms depending on the consequence maintaining the behavior. Attention extinction may occur if attention from caregivers or others is withheld contingent on the occurrence of the target behavior. Note that ignoring an individual's problem behavior will be effective in reducing that behavior only when that behavior is maintained by contingent social attention. Extinction of escape/avoidance behavior consists of preventing the individual from escaping from or avoiding the event that elicits negatively reinforced problem behavior. Finally, in case of automatic positive reinforcement, sensory extinction involves removing the sensory consequences resulting from problem behavior.

Extinction is associated with several negative side effects. A clinically important side effect of extinction is the occurrence of the so-called extinction burst, which is an increase, albeit temporary, in the frequency or intensity of the target behavior at the beginning of the treatment (see Weiskop, Richdale, & Matthews (2005) for an extinction burst during treatment of attention-maintained sleep disruptive behavior). Extinction is also accompanied by increased behavioral variability, as seen by the emergence of novel behaviors or the reemergence of old behaviors. This can be the basis of reinforcement of appropriate behavior. Clinicians should be aware that an extinction burst is a signal of treatment effectiveness, not ineffectiveness. EXT in combination with differential reinforcement often results in a faster reduction in the target behavior than EXT used alone (Petscher, Rey, & Bailey, 2009; Seligson Petscher & Bailey, 2008) and is therefore preferable to EXT alone.

IV. Summary

Vicarious extinction is theoretically based on Albert Bandura's learning through modeling paradigm. It is an analogous process to in vivo extinction, the difference being that the former occurs when an observer sees a model performing a behavior that is no longer reinforced, rather than not being reinforced (extinguished) directly. It was shown through research conducted in the 1960s and early 1970s to be an effective technique for extinguishing fear and avoidance behavior. It is more flexible and adaptable than in vivo extinction because the observer does not have to experience the extinction directly. However, live modeling combined with guided participation, which is essentially a combination of vicarious and in vivo extinction, was found to be more effective than other forms of vicarious extinction alone.