The prefrontal-thalamic axis and classical conditioning

Using the New Zealand albino rabbit as an animal model, it has been determined that the medial prefrontal cortex is intimately involved in, and may be necessary for, acquisition of classically conditioned bradycardia. The interconnected nuclei of the medial thalamus, most notably the mediodorsal nucleus, conversely appear to be associated with the development of the tachycardia that accompanies classical conditioning of the eyeblink and nictitating membrane response. Neuroanatomical, electrophysiological, brain stimulation, and lesion data are reviewed, which support the conclusion that the prefrontal-thalamic axis is intimately involved in acquisition of classically conditioned visceral changes.     

The acquisition of conditioned suppression as a function of interstimulus interval duration

The effect of interstimulus interval (ISI) variation on the acquisition of a classically conditioned emotional response was investigated using a one-trial conditioning procedure. The optimum ISI was found to be 10 s with a bidirectional gradient for conditioned suppression at ISI above and below 10 s. Control groups demonstrated that conditioning was not a function of either pseudoconditioning, sensitization or stimulus novelty.     

Extinction of a classically conditioned G.S.R. as a function of awareness

Insight, defined as the ability to verbalize the stimulus-response contingency of acquisition has bsen found to be unrelated to the extinction of an operantly conditioned verbal response. In the present study, the authors classically conditioned a GSR response in thirty Ss without insight, divided them into insight and no-insight groups following acquisition, and ran extinction trials which demonstrated a positive relationship (P< 0.001) under these particular conditions. It was concluded that the role of insight in extinction is a complex function of mode of conditioning (classical vs. operant) and/or mode of response (voluntary vs. involuntary), and that further studies utilizing the classical-voluntary and operant-involuntary combinations will be needed to define that function.     

Facilitated Acquisition of the Classically Conditioned Eyeblink Response in Male Rats After Systemic IL-1?

Exposure to inescapable stressors enhances cue-dependent learning in male rats; enhanced learning is apparent as facilitated acquisition of the classically conditioned eyeblink respouse (CCER). The proinflammatory cytokines, in particular interleukin (IL)-1beta, are presumed to orchestrate a number of acute-phase stress responses in rats, most notably fever, reduced feeding, and inactivity. Little is known of the impact proinflammatory cytokines have on learning and memory processes. Here, we address the effects of IL-1beta treatment on acquisition of the classically conditioned eyeblink response 2 hours [?] after injection in male rats. Training was accomplished with a delay-type paradigm (500-ms conditional stimulus coterminating with a 10-ms periorbital unconditional stimulation). Facilitated acquisition was clearly apparent in rats treated with IL-1beta (3.0 microg/kg). In a second experiment, we compared rats treated with 3.0 microg/kg to those treated with 1.0 microg/kg. Facilitated acquisition was reproduced, but the lower dose did not appreciably affect acquisition. These data further support contentions that IL-1beta has anxiogenic properties, affecting basic new motor learning in a manner similar to that observed after exposure to stress.     

Facilitated acquisition of the classically conditioned eyeblink response in male rats after systemic IL-1β

Exposure to inescapable stressors enhances cue—dependent learning in male rats; enhanced learning is apparent as facilitated acquisition of the classically conditioned eyeblink respouse (CCER). The proinflammatory cytokines, in particular interleukin (IL)-1β, are presumed to orchestrate a number of acute-phase stress responses in rats, most notably fever, reduced feeding, and inactivity. Little is known of the impact proinflammatory cytokines have on learning and memory processes. Here, we address the effects of IL-1β treatment on acquisition of the classically conditioned eyeblink response 2 hours [?] after injection in male rats. Training was accomplished with a delay-type paradigm (500-ms conditional stimulus coterminating with a 10-ms periorbital unconditional stimulation). Facilitated acquisition was clearly apparent in rats treated with IL-1β (3.0 μg/kg). In a second experiment, we compared rats treated with 3.0 μg/kg to those treated with 1.0 μg/kg. Facilitated acquisition was reproduced, but the lower dose did not appreciably affect acquisition. These data further support contentions that IL-1β has anxiogenic properties, affecting basic new motor learning in a manner similar to that observed after exposure to stress.     

Rapid associative learning: Conditioned bradycardia and its central nervous system substrates

It has become clear from the study of different response systems during classical conditioning that some responses are acquired quite rapidly and others show a much slower rate of acquisition. The most often studied rapidly acquired responses have been classically conditioned autonomic changes (e.g., heart rate); the slowly acquired responses most often studied are skeletal responses, such as the eyeblink or leg flexion response. Although there are various other differences between rapidly acquired and slowly acquired responses, we have suggested that the most important difference is the possibility that they represent different stages of the learning process. In the present review I describe research in our laboratory that has focused on conditioned bradycardia as a model system of a rapidly acquired associative system and contrast it with the more slowly acquired Pavlovian conditioned eyeblink response. I also describe the generality of conditioned bradycardia and discuss the differential role of subdivisions of the prefrontal cortex as a substrate for mediating this response. Finally, I briefly discuss the other brain areas involved in conditioned bradycardia, and its functional significance as it relates to the learning process.     

Cerebellar lesions abolish an avoidance response in rabbit

Unilateral lesions of the cerebellar dentate-interpositus (DI) nuclei abolished or impaired the reacquisition of a nictitating membrane (NM) conditioned avoidance response (CAR) in rabbits. Animals that sustained damage to the DI nuclei and surrounding areas displayed little or no reacquisition of the NM-CAR ipsilateral to the lesion, but displayed excellent acquisition of the response when training was shifted to the contralateral eye. These results agree with previous investigations of the effects of cerebellar lesions on the classically conditioned NM and leg-flexion responses. The DI nuclei appear to be essential for the retention of conditioned responses in a variety of associative paradigms.     

Limbic networks and associative learning: II. Entorhinal contributions to spontaneous alternation,passive avoidance,and taste aversion learning

The hippocampus plays an important role in learning and memory, but the precise nature of that involvement remains uncertain. Transection of the perforant path, a primary input pathway to the hippocampus, has been shown to produce changes in reaction to novelty and acquisition of active avoidance; the nature and magnitude of these changes vary with lateral or medial perforant path damage. In a series of experiments on adult rats, the role of these pathways in spontaneous alternation, exploration, acquisition and extinction of conditioned responses, passive avoidance, and conditioned taste aversion was investigated. Lateral transection reduced exploration while medial transection facilitated acquisition of an active avoidance response; no effects were observed on any other measure. Results are discussed in terms of what perforant path damage might reveal regarding the interactions of the hippocampus with other brain regions.     

Limbic networks and associative learning: II. Entorhinal contributions to spontaneous alternation, passive avoidance, and taste aversion learning

The hippocampus plays an important role in learning and memory, but the precise nature of that involvement remains uncertain. Transection of the perforant path, a primary input pathway to the hippocampus, has been shown to produce changes in reaction to novelty and acquisition of active avoidance; the nature and magnitude of these changes vary with lateral or medial perforant path damage. In a series of experiments on adult rats, the role of these pathways in spontaneous alternation, exploration, acquisition and extinction of conditioned responses, passive avoidance, and conditioned taste aversion was investigated. Lateral transection reduced exploration while medial transection facilitated acquisition of an active avoidance response; no effects were observed on any other measure. Results are discussed in terms of what perforant path damage might reveal regarding the interactions of the hippocampus with other brain regions.     

Is the medial amygdala part of the neural circuit modulating conditioned defeat in Syrian hamsters?

Conditioned defeat is a model wherein hamsters that have previously experienced a single social defeat subsequently exhibit heightened levels of avoidance and submission in response to a smaller, non-aggressive intruder. While we have previously demonstrated the critical involvement of the basolateral and central nuclei of the amygdala in the acquisition and expression of conditioned defeat, the role of the medial amygdala has yet to be investigated. In Experiment 1, muscimol, a GABA(A) receptor agonist, was infused bilaterally into the MeA prior to initial defeat training. Experiment 2 examined the effects of muscimol injections given prior to subsequent testing with a non-aggressive intruder. Finally, in Experiment 3, anisomycin was used to block protein synthesis in the medial and basolateral amygdala to examine the involvement of these nuclei in memory consolidation related to conditioned defeat. Submissive behavior was significantly reduced in animals that received muscimol prior to initial defeat training as well as in animals injected prior to testing with the non-aggressive intruder, indicating that the MeA is necessary for the acquisition and expression of conditioned defeat. In Experiment 3, however, anisomycin reduced conditioned defeat only when administered into the BLA, and not when injected into the MeA. The results of the present series of experiments suggest that, while the MeA may serve an important gateway for sensory information that is crucial for conditioned defeat, it does not appear to play a role in the plasticity including this behavioral response to social defeat.     

Differential effects of progesterone and medroxyprogesterone on delay eyeblink conditioning in ovariectomized rats

Ovarian hormones modulate acquisition processes involved in classical conditioning. Although progesterone has been indirectly implicated, its role in classical conditioning of the eyeblink response has not been directly investigated. We assessed the effects of daily dosing of progesterone or medroxyprogesterone (MPA), a non-metabolized synthetic progestin, upon the acquisition of a classically conditioned eyeblink response in ovariectomized (OVX) female rats. Rats were dosed 4h prior to each training session with 0.1 or 1.5 mg/kg of either of these hormones or sesame oil. A delay conditioning paradigm was employed using a 500 ms conditioned stimulus coterminating with a 10 ms 10 V unconditioned stimulus. At the low dose, progesterone and MPA rats did differ from each other, with MPA-treated rats learning slower, but neither group differed from OVX-oil or Sham-oil controls. No group differences in acquisition were observed at the higher dose. During extinction trials, high-dose MPA-treatment and OVX-oil groups extinguished quicker than the high-dose progesterone-treated group. In addition, unconditional response (UR) amplitudes were lower in all OVX groups, regardless of hormone or oil treatment, compared to the sham-oil group. Since MPA did not affect extinction, it is likely the slower extinction in the progesterone-treated rats is due to a metabolite of progesterone. Corticosterone is discussed as a likely candidate for such a role. In addition, we found chronic absence of ovarian hormones decreased UR amplitudes, although differences in UR amplitudes were not associated with changes in the acquisition process. These results are discussed with respect to differences in the hormonal effects upon acquisition versus extinction processes and how these data may explain reports of learning differences in women based on oral contraceptive usage.     

Classical conditioning of beginning reading responses

The present study describes the first demonstration that laboratory-controlled experimental procedures can lead to the successful acquisition and subsequent retention of classically conditioned beginning reading responses (CCBRRs) in children of both sexes and mean age of 4 yr. Anticipatory instructions combined with higher-order classical conditioning temporally arranged into a trace conditioning paradigm presented for 10 trials for each response to be learned led to beginning reading responses being successfully acquired by 20 children during 95% of the 2,220 total acquisition learning trials and subsequently correctly recalled on 114 of the 222 retention test trials. Findings support the view that perhaps the relatively sudden and sustained acquisition learning curves for reading responses on the second-signalling-system level of behavior in the present study may be quite different from the relatively slow and incremental learning curves usually obtained in classical conditioning of the autonomic type which occur on the first-signalling-system level.     

Animal learning of independent and correlated events

Evidence suggests that experience of a truly random presentation of a conditioned and unconditioned stimulus (CS and US) retards subsequent acquisition of a classically conditioned response when the CS and US are paired. In these studies, no events reliably predict the US during the first part of training. The present experiment demonstrated impairment of subsequent appetitive conditioning in rats following a random CS-US presentation, both when another predictive cue is present and when it is absent during this experience. Results are discussed with reference to “general attention” theory, and learned helplessness. However, the data support an interpretation in terms of learning a specific relationship between a particular CS and US.     

Microstimulation reveals opposing influences of prelimbic and infralimbic cortex on the expression of conditioned fear

Recent studies using lesion, infusion, and unit-recording techniques suggest that the infralimbic (IL) subregion of medial prefrontal cortex (mPFC) is necessary for the inhibition of conditioned fear following extinction. Brief microstimulation of IL paired with conditioned tones, designed to mimic neuronal tone responses, reduces the expression of conditioned fear to the tone. In the present study we used microstimulation to investigate the role of additional mPFC subregions: the prelimbic (PL), dorsal anterior cingulate (ACd), and medial precentral (PrCm) cortices in the expression and extinction of conditioned fear. These are tone-responsive areas that have been implicated in both acquisition and extinction of conditioned fear. In contrast to IL, microstimulation of PL increased the expression of conditioned fear and prevented extinction. Microstimulation of ACd and PrCm had no effect. Under low-footshock conditions (to avoid ceiling levels of freezing), microstimulation of PL and IL had opposite effects, respectively increasing and decreasing freezing to the conditioned tone. We suggest that PL excites amygdala output and IL inhibits amygdala output, providing a mechanism for bidirectional modulation of fear expression.     

Effect of intermittent reinforcement on acquisition and retention in delayed matching-to-sample in pigeons

Experiments 1 and 2 involved independent groups that received primary reinforcement after a correct match with a probability of 1.0, .50 or .25. Correct matches that did not produce primary reinforcement produced a conditioned reinforcer. Both experiments revealed little evidence that acquisition or retention was adversely affected by use of intermittent reinforcement. Experiment 3 involved a group that received 100% reinforcement and two others that received 25% reinforcement, one of which received conditioned reinforcement and the other did not. Following acquisition and retention testing, birds in group 100% and group 25% with conditioned reinforcement were exposed to 25% reinforcement and no conditioned reinforcement. Results revealed that conditioned reinforcement was important in promoting acquisition but was irrelevant in maintaining performance. It was concluded that intermittent reinforcement, especially when combined with conditioned reinforcement during acquisition, supports levels of acquisition and retention comparable to that of continuous reinforcement. Theoretically, the findings are consistent with an extension of Blough's instance-based theory of discrimination performance and, practically, they suggest that use of intermittent reinforcement could result in increased efficiency and economy in labs using delayed matching.     

Damage to the lateral and central, but not other, amygdaloid nuclei prevents the acquisition of auditory fear conditioning

It is well established that the amygdala plays an essential role in Pavlovian fear conditioning, with the lateral nucleus serving as the interface with sensory systems that transmit the conditioned stimulus and the central nucleus as the link with motor regions that control conditioned fear responses. The lateral nucleus connects with the central nucleus directly and by way of several other amygdala regions, including the basal, accessory basal, and medial nuclei. To determine which of these regions is necessary, and thus whether conditioning requires the direct or one of the indirect intra-amygdala pathways, we made lesions in rats of the lateral, central, basal, accessory basal, and medial nuclei, as well as combined lesions of the basal and accessory basal nuclei and of the entire amygdala. Animals subsequently underwent fear conditioning trials in which an auditory conditioned stimulus was paired with a footshock unconditioned stimulus. Animals that received lesions of the lateral or central nucleus, or of the entire amygdala, were dramatically impaired, whereas the other lesions had little effect. These findings show that only the lateral and central nuclei are necessary for the acquisition of conditioned fear response to an auditory conditioned stimulus.     

Medial auditory thalamus inactivation prevents acquisition and retention of eyeblink conditioning

The auditory conditioned stimulus (CS) pathway that is necessary for delay eyeblink conditioning was investigated using reversible inactivation of the medial auditory thalamic nuclei (MATN) consisting of the medial division of the medial geniculate (MGm), suprageniculate (SG), and posterior intralaminar nucleus (PIN). Rats were given saline or muscimol infusions into the MATN contralateral to the trained eye before each of four conditioning sessions with an auditory CS. Rats were then given four additional sessions without infusions to assess savings from the initial training. All rats were then given a retention test with a muscimol infusion followed by a recovery session. Muscimol infusions through cannula placements within 0.5 mm of the MGm prevented acquisition of eyeblink conditioned responses (CRs) and also blocked CR retention. Cannula placements more than 0.5 mm from the MATN did not completely block CR acquisition and had a partial effect on CR retention. The primary and secondary effects of MATN inactivation were examined with 2-deoxy-glucose (2-DG) autoradiography. Differences in 2-DG uptake in the auditory thalamus were consistent with the cannula placements and behavioral results. Differences in 2-DG uptake were found between groups in the ipsilateral auditory cortex, basilar pontine nuclei, and inferior colliculus. Results from this experiment indicate that the MATN contralateral to the trained eye and its projection to the pontine nuclei are necessary for acquisition and retention of eyeblink CRs to an auditory CS.     

Intra-medial prefrontal administration of SCH-23390 attenuates ERK phosphorylation and long-term memory for trace fear conditioning in rats

The prefrontal cortex is known to be involved in the acquisition of trace conditioning, a higher-cognitive form of Pavlovian conditioning in which a conditioned stimulus and an unconditioned stimulus are separated by a time gap. We have recently reported that medial prefrontal (mPFC) extracellular-signal regulated kinase (Erk) phosphorylation is involved in the long-term memory storage of trace fear conditioning. Because of the important role dopamine D1 receptors play in prefrontal function, such as working memory, and due to evidence that dopamine D1 receptor activity can modulate plasticity, we investigated their role in prefrontal Erk phosphorylation following trace fear conditioning. We found that inhibition of dopamine D1 receptors through intra-mPFC infusion of SCH-23390 (1 microg/0.5 microL) 15 min prior to trace fear conditioning resulted in a decrease in training-related Erk phosphorylation. Additionally, pre-training intra-mPFC infusion of SCH-23390 also resulted in the impairment of long-term retention of CS-US association. These findings implicate mPFC dopamine D1 receptor activity in the storage of long-term memory for higher-cognitive associative tasks.     

Retention of a classically conditioned reflex response in spinal cat

Retention of classically conditioned flexion reflex facilitation was examined in unanesthetized, decerebrate, acute spinal cats. Flexion reflex facilitation, recorded from the tibialis anterior muscle, was obtained by pairing saphenous nerve stimulation (the conditioned stimulus) with superficial peroneal nerve stimulation (the unconditioned stimulus). The flexion reflex declined in control animals receiving the same number of nerve stimuli over the same time span, but in an explicitly unpaired sequence. To investigate retention, conditioned stimuli were presented at 5-min intervals following acquisition for a 2 1/2-h period. During this time a significant difference between conditioning and control groups was maintained even to the last trial, with no indication that the difference was subsiding over time. The results support the possibility that a classical conditioning paradigm applied to the spinal cord can induce alterations in spinal reflexes of long duration. Furthermore, the results appear to rule out post-tetanic potentiation as a mechanism producing the observed long-term effects.     

Limbic networks and associative learning: I. Entorhinal contributions to instrumental conditioning

The hippocampal formation is a highly delineated brain structure that is believed to play a prominent role in learning and memory. The present experiment evaluated the contributions of medial and lateral perforant path input to bar press-conditioning under (a) continuous and (b) differential reinforcement of low rates of responding (DRL) schedules, and (c) shuttlebox avoidance conditioning. Bilateral deafferentation of either pathway had no effect on the acquisition of bar press responses or on performance under the DRL schedule. Deafferentation of the medial pathway facilitated acquisition of avoidance responses in a manner much like the effects seen in hippocampectomized animals. It is suggested that the medial perforant path participates in the expression of correlated patterns of neuronal discharge known to develop within the hippocampus and that this “model” serves to modulate the temporal characteristics of simple conditioned reflexes. Loss of the modulatory influence of the model may affect acquisition and extinction rates. Contributions of other hippocampal circuits are discussed in relation to established deficits. Preliminary results of this experiment were presented at the second annual convention of the American Psychological Society, Dallas, Texas, June 1990.