Spatio-temporal dynamics of brain mechanisms in aversive classical conditioning: high-density event-related potential and brain electrical tomography analyses

Social cognition, including complex social judgments and attitudes, is shaped by individual learning experiences, where affect often plays a critical role. Aversive classical conditioning-a form of associative learning involving a relationship between a neutral event (conditioned stimulus, CS) and an aversive event (unconditioned stimulus, US)-represents a well-controlled paradigm to study how the acquisition of socially relevant knowledge influences behavior and the brain. Unraveling the temporal unfolding of brain mechanisms involved appears critical for an initial understanding about how social cognition operates. Here, 128-channel ERPs were recorded in 50 subjects during the acquisition phase of a differential aversive classical conditioning paradigm. The CS+ (two fearful faces) were paired 50% of the time with an aversive noise (CS upward arrow + /Paired), whereas in the remaining 50% they were not (CS upward arrow + /Unpaired); the CS- (two different fearful faces) were never paired with the noise. Scalp ERP analyses revealed differences between CS upward arrow + /Unpaired and CS- as early as approximately 120 ms post-stimulus. Tomographic source localization analyses revealed early activation modulated by the CS+ in the ventral visual pathway (e.g. fusiform gyrus, approximately 120 ms), right middle frontal gyrus (approximately 176 ms), and precuneus (approximately 240 ms). At approximately 120 ms, the CS- elicited increased activation in the left insula and left middle frontal gyrus. These findings not only confirm a critical role of prefrontal, insular, and precuneus regions in aversive conditioning, but they also suggest that biologically and socially salient information modulates activation at early stages of the information processing flow, and thus furnish initial insight about how affect and social judgments operate.     

Effects of expectation on amygdala-kindled convulsions and interictal behaviour in rats: confirmation of a Pavlovian mechanism

We found evidence that the effects of exposure to the stimulation environment on the convulsions and interictal behaviour of basolateral amygdala (BA)-kindled rats are the result of Pavlovian conditioning. In Experiment 1, the rats first received 45 BA stimuli in one environment (CS+) and 45 sham stimuli in another environment (CS-). During this first phase of the experiment, the CS+ and CS- began to exert differential effects on interictal behaviour and convulsions. Then, during the second phase, rats received 45 stimuli and 45 sham stimuli, either as before in their original CS+ and CS- (nonreversal rats) or with their original CS+ and CS- interchanged (reversal rats). During the second phase, the CS+ and CS- continued to exert the same pattern of differential effects on the interictal behaviour and convulsions of the nonreversal rats, whereas the reversal rats began to display the reverse pattern. In Experiment 2, rats received 60 sham stimuli in one environment (CS1). Then, half of the rats received 30 stimuli in a second environment (CS2) and 30 sham stimuli in CS1 (no-pre-exposure rats), whereas the other half (pre-exposure rats) received 30 stimuli and 30 sham stimuli in CS1. Pre-exposure to the stimulation environment attenuated the kindled convulsions. These findings confirm that the conditioned effects of kindling are mediated by a Pavlovian mechanism.     

Dynamic neural activity recorded from human amygdala during fear conditioning using magnetoencephalography

Magnetoencephalography (MEG) was used to record the dynamics of amygdala neuronal population activity during fear conditioning in human participants. Activation during conditioning training was compared to habituation and extinction sessions. Conditioned stimuli (CS) were visually presented geometric figures, and unconditioned stimuli (US) were aversive white-noise bursts. The CS+ was paired with the US on 50% of presentations and the CS- was never paired. The precise temporal resolution of MEG allowed us to address the issue of whether the amygdala responds to the onset or offset of the CS+, and/or the expectation of the initiation or offset of the an omitted auditory US. Fear conditioning elicited differential amygdala activation for the unpaired CS+ compared to the CS-, extinction and habituation. This was especially robust in the right hemisphere at CS onset. The strongest peaks of amygdala activity occurred at an average of 270 ms in the right and 306 ms in the left hemisphere following unpaired CS+ onset, and following offset at 21 ms in the left and 161 ms in the right (corresponding to an interval of 108 ms and 248 ms after the anticipated onset of the US, respectively). However, the earliest peaks in this epoch preceded US onset in most subjects. Thus, the activity dynamics suggest that the amygdala both differentially responds to stimuli and anticipates the arrival of stimuli based on prior learning of contingencies. The amygdala also shows stimulus omission-related activation that could potentially provide feedback about experienced stimulus contingencies to modify future responding during learning and extinction.     

The neural representation of odor is modulated by the presence of a trigeminal stimulus during odor encoding.

OBJECTIVES: Odor perception does not simply consist in hierarchical processing from transduction to a single "true" cerebral representation. Odor sensation may be modulated by available sensory information during encoding. The present study set out to examine whether the presence of a pure trigeminal stimulus during odor encoding may modulate odor perception at both behavioral and cortical levels. METHODS: Participants were tested in a 2-session within-subject design: first, an odor encoding session included a delay conditioning procedure in which relatively selective olfactory stimulants (phenyl ethyl alcohol or vanillin, Conditioned Stimulus+, CS+) were presented either with a pulse of CO(2) (Unconditioned Stimulus, US), or alone (Conditioned Stimulus-, CS-); then, in the second session, both pure odorants (CS+ and CS-) were presented alone. During this second session, olfactory event-related potentials were simultaneously recorded and analyzed at different electrode sites including Cz and Pz (sites known to have maximal amplitudes for trigeminal and olfactory stimuli, respectively). After each trial, subjects were asked to rate odor intensity and hedonics. RESULTS: The results showed that CS+ intensity ratings increased in 8 subjects and decreased in 6. Cortically, a group effect was observed for P2 amplitude, which increased in the "CS+ intensity increase" group vs. the "CS+ intensity decrease" group at Cz (p<0.05) but not at Pz (p>0.05). CONCLUSIONS: This result suggests that the presence of a pure trigeminal stimulus (CO(2)) during odor encoding alters the neural representation of a pure odor. SIGNIFICANCE: The neural representation of odors comprises not only the odor itself but also contextual information (trigeminal in the present case) presented during encoding.     

Evaluative differential conditioning of disgust: a sticky form of relational learning that is resistant to extinction

The present study sought to (a) test whether autonomic (i.e., electrodermal) and evaluative conditioning can be differentially established to verbal CSs, and (b) whether extinction procedures can reliably attenuate differential conditioned evaluative responding. Thirty undergraduates underwent a 10-min adaptation period followed by three consecutive conditioning phases: habituation, acquisition, and extinction. Conditioning involved participants viewing two semi-randomly presented words on a computer monitor. During acquisition, one word (CS+) was reliably paired 12 times with the UCS (pictorial stimuli depicting bodily mutilation), whereas the remaining word (CS-) was presented 12 times and reliably followed by neutral pictures (inanimate common objects). As predicted, electrodermal and evaluative responses during acquisition were of larger magnitude to the CS+ compared to the CS-. During extinction, participants continued to evaluate the CS+ as more disgusting relative to the CS-, whereas distress and fear-related emotional ratings attenuated across extinction trials. The implications of these findings for the modifiability of disgust-based evaluative responses in specific anxiety disorders will be discussed.     

Evaluative conditioning is intact after unilateral resection of the anterior temporal lobe in humans

Several lesion and functional imaging studies conducted in animals and humans suggest that structures within the amygdaloid nuclear complex (ANC) are important for the occurrence of fear conditioning. Whether this brain structure is also critical for evaluative conditioning, has been investigated less frequently. In the current experiment, a group of participants with unilateral resection of the anterior temporal lobe and a control group received a differential evaluative flavor-taste conditioning task. In the pre-acquisition phase, two fruit flavors (the conditioned stimuli (CSs)) were presented and participants were instructed to evaluate both. In the subsequent acquisition phase, one of these fruit flavors (CS+) was presented together with a bad tasting substance Tween20 (polysorbate 20, the US), while the other flavor (CS-) was never paired with Tween20. Finally, in the post-acquisition phase, the two flavors were presented again without Tween20 and participants were asked to evaluate both of them for a last time. The control group as well as the lesion group rated the CS+ in the post-acquisition phase less favorable than in the pre-acquisition phase, while the ratings of the CS- remained the same in both phases. We clearly demonstrated evaluative conditioning in both test groups. Because the lesion group had still one intact ANC it would be premature, however, to conclude that the ANC is not involved in evaluative conditioning. We conclude that despite evidence for impaired fear conditioning, unilateral damage to the ANC does not impair evaluative conditioning.     

Psychophysiologic assessment of aversive conditioning in posttraumatic stress disorder.

BACKGROUND: The objective of this study was to evaluate the acquisition, generalization, and extinction of conditioned physiologic responses to aversive stimuli in posttraumatic stress disorder (PTSD). METHODS: Thirty-six PTSD patients, 20 individuals with past trauma and no current PTSD, and 30 mentally healthy individuals without exposure to major trauma underwent a differential aversive conditioning experiment. Bursts of 105 dB white noise were used as unconditioned stimuli (UCSs), and 35x24 mm slides of different colors served as either CS+ (paired) or CS- (unpaired) stimuli. Heart rate (HR) and nondominant palm skin conductance (SC) were measured at rest and between 1 and 4 sec following each CS presentation. RESULTS: The PTSD group showed higher levels of resting SC and resting HR, larger SC responses to the initial presentation of unpaired CSs, larger HR responses following paired CS+ stimuli, larger SC responses to unpaired CS- during acquisition and extinction, and larger SC and HR responses to CS+ during extinction. The group differences in responses to CS+ during extinction remained statistically significant after controlling for age, resting physiologic levels, and initial responsivity. CONCLUSIONS: PTSD is associated with elevated autonomic responses to both innocuous and aversive stimuli, with larger responses to unpaired cues and with reduced extinction of conditioned responses.     

Cortical evoked potential changes during classical conditioning of morphine dependence in rats

A classical conditioning experiment was designed to determine if a conditioned neural response would develop and persist in cortical evoked potentials elicited by a foreleg stimulus (CS+) that was paired with morphine administration during the development of dependence and subsequent withdrawal. A stimulus to the other foreleg (CS-) was presented explicitly unpaired with morphine delivery. After dependence was established, the rats were taken from the experimental chamber and withdrawn from morphine for 6 days in their home cages. Finally, during the testing phase, the animals were returned to the experimental chamber and the foreleg stimuli were presented. The CS+ was paired with either morphine or saline injections. Changes due to both morphine effects only and conditioning were observed. The conditioned response, however, was present only in the cortical evoked potentials recorded from those animals receiving contralateral foreleg stimulation as the CS+. The conditioned neural response persisted after withdrawal and was present in both the drug-free and morphine-intoxicated animals. These results provide support for the relapse theory that a nonextinguished conditioned response is retained after withdrawal. However, further experiments are necessary to determine if these conditioned responses can elicit drug-seeking behavior.     

Disruption of bradycardia associated with discriminative conditioning in combat veterans with PTSD

The effects of combat-related posttraumatic stress disorder (PTSD) on heart rate (HR) responding associated with a discriminative delay eyeblink (EB) conditioning paradigm are reported. Combat PTSD+, Combat PTSD-, and Noncombat PTSD- veterans were assessed with psychometric self-report measures, and baseline heart rate variability (HRV) was measured before receiving a 72-trial session of discriminative EB classical conditioning. Two types (red or green light) of conditioned stimuli (CS) were used: one (CS+) predicted a tone, followed immediately by an aversive stimulus (corneal airpuff); the other (CS-) predicted a tone alone, not followed by the airpuff. The light signal was presented for 5 seconds, during which HR was measured. On all psychometric measures, the PTSD+ subgroup was significantly different from the PTSD- subgroups (Combat + Noncombat), and the PTSD- subgroups did not significantly differ from each other. A linear deceleration in HR to CS+ and CS- signals was found in the combined PTSD- subgroup and on CS- trials in the PTSD+ subgroup, but was not present on CS+ trials in the PTSD+ subgroup. Results are interpreted with respect to a behavioral stages model of conditioned bradycardia and in terms of neural substrates which are both critical to HR conditioning and known to be abnormal in PTSD.     

Effects of interstimulus interval and contingency on classical conditioning of the Aplysia siphon withdrawal reflex

The siphon withdrawal reflex of Aplysia undergoes differential classical conditioning with cutaneous stimulation of the siphon or mantle shelf as the discriminative conditioned stimuli (CS+ and CS-) and shock to the tail as the unconditioned stimulus (US). The reflex has proved to be useful for analyzing the neural mechanisms of conditioning. To test the generality of this experimental system, we have begun to compare the properties of conditioning in Aplysia with those of conditioning in vertebrates. We first examined the effect of the interstimulus interval (ISI) by varying the time between presentation of the CS+ and the US in different groups of animals. Significant differential conditioning was obtained when the onset of the CS+ preceded the onset of the US by 0.5 sec, and marginal conditioning was obtained when the ISI was 1.0 sec. By contrast, no significant conditioning occurred when the CS+ preceded the US by 2, 5, or 10 sec, when the onsets of the stimuli were simultaneous, or when US onset preceded the CS+ by 0.5, 1.0, or 1.5 sec (backward conditioning). We next examined the effect of contingency by giving one group of animals normal differential conditioning, and a second group the same training but with additional USs inserted between the paired trials. Presentation of these additional USs reduced the degree to which the US was contingent on the CS+, but did not change the number of pairings. Animals receiving normal training again showed significant conditioning, whereas animals receiving additional USs showed no conditioning.(ABSTRACT TRUNCATED AT 250 WORDS)     

Discriminative delay Pavlovian eyeblink conditioning in veterans with and without posttraumatic stress disorder

BACKGROUND: Impaired eyeblink (EB) classical conditioning using a delay paradigm has previously been shown in combat veterans, as well as in a group of depressed adults, compared to normal individuals. Significant deficits in immediate memory (IM) in combat PTSD+ veterans, compared to normal controls, have also been previously shown, but these differences became non-significant after controlling for level of self-reported depression. Furthermore, EB conditioning has been shown to be significantly correlated with heart rate variability (HRV) in normal adults. The present study examined how depression (self-reported), IM, and resting HRV are related to discriminative delay classical EB conditioning in veterans with and without PTSD. METHOD: Three groups of subjects (combat PTSD+, combat PTSD-, and non-combat PTSD-) were assessed for self-report of depression and anxiety, as well as IM and HRV. Subjects received a single session of discriminative EB classical conditioning in which the conditioned stimulus (CS) was a light signal (either red or green) compounded with a tone. On CS+ trials, the light-tone compound stimulus co-terminated with a corneal airpuff (unconditioned stimulus, US), thus producing a delay paradigm. On CS- trials the appropriate light-tone stimulus was presented but not followed by the airpuff US. EB amplitude and frequency were recorded. RESULTS: PTSD+ subjects had greater self-reported depression and anxiety scores than the two control groups, as well as lower scores on a measure of IM. However, the IM difference was not significant after the effects of self-reported depression and anxiety were controlled. EB CR amplitude was significantly greater to CS+ than CS- for all three groups. EB amplitude to both the US (airpuff) and the CS+ declined over trials, but was significantly lower in the combat PTSD+ group compared to the combined PTSD- groups. Subjects who reached an EB CR acquisition criterion had significantly greater scores on IM than those who did not reach criterion. Factor analysis of the entire data set revealed four factors corresponding to (1) self-reported depression and anxiety, (2) IM, (3) HRV, and (4) EB amplitude. EB frequency was significantly predicted by IM and HRV. CONCLUSIONS: These data extend our previous results by showing deficits in EB conditioning among combat PTSD+ veterans that were associated with lower IM and resting HRV, but were not associated with self-report of depression.     

Dopamine efflux in nucleus accumbens shell and core in response to appetitive classical conditioning

Dopamine transmission within the nucleus accumbens has been implicated in associative reinforcement learning. We investigated the effect of appetitive classical conditioning on dopamine efflux in the rat nucleus accumbens shell and core, as dopamine may be differentially activated by conditioned and unconditioned stimuli (CS, US) in these subregions. After implantation of microdialysis cannulae, rats were food restricted and trained for three consecutive days with three acquisition sessions per day. A 10-s noise (CS) was immediately followed by the delivery of two reward pellets (US) for the conditioned group (paired presentation), whereas conditioned stimuli and unconditioned stimuli were presented at random for the control group (unpaired presentation). On the fourth day, all rats were given a further CS + US session and two CS-alone sessions, and extracellular dopamine concentrations were measured (7.5 min/per sample). Behavioural measures (number of nose pokes, latency to nose poke after conditioned stimuli onset, locomotor activity) demonstrated that the paired groups showed a high level of conditioning. CS + US presentation increased dopamine equally in both shell and core of the paired and unpaired groups. CS alone presentation induced a conditioned dopamine release only in the paired groups. No significant difference was found between shell and core. Unlike previous conditioning paradigms involving either a more salient US (foot shock, addictive drug) or a more complex CS, the present paradigm, using normal reward pellets as US and a discrete auditory stimulus as CS, did not lead to differential responses in dopamine efflux in shell and core subregions of the nucleus accumbens.     

Discriminative long-term retention of rapidly induced multiunit changes in the hippocampus, medial geniculate and auditory cortex

Multiunit activity was chronically recorded in the hippocampus (CA3 field), the magnocellular medial geniculate (MGm) and the auditory cortex (AC) of rats during acquisition (12 daily sessions, 10 trials per session) and long-term retention of differential classical conditioning (tones paired with footshocks). Marked increases of multiunit discharges to CS + presentations were first detected in the MGm (5-10 trials) followed (10-20 trials) by the emergence of discriminative responses in the hippocampus and in the AC. During long-term retention tests, 45 days after the end of conditioning, CS + selective responses were observed in the 3 structures. We propose that learning-induced changes in the conditioned stimulus (CS) sensory pathway can have the same temporal stability as the sensory plasticity observed during development or post injury in adult animals.     

Impaired fear conditioning in Alzheimer's disease

Classical conditioning of the fear response is a basic form of nondeclarative (nonconscious) memory that mediates both normal and pathological responses to aversive stimuli. Because fear conditioning critically depends on the amygdala, a medial temporal lobe structure that frequently undergoes significant pathological changes early in the course of Alzheimer's disease (AD), we hypothesized that fear conditioning would be impaired in patients with mild to moderate AD. We examined simple classical fear conditioning in a group of 10 patients with probable AD and 14 demographically matched, neurologically intact elderly controls. During conditioning, one stimulus (e.g. a green rectangle, the conditioned stimulus (CS+)), was paired with an aversive stimulus (a loud noise, the unconditioned stimulus (US)) using a partial reinforcement conditioning schedule. The opponent color (e.g. red rectangle), the CS-, was never paired with the US. The elderly controls acquired robust fear responses as demonstrated by their differential skin-conductance responses to the CS+ and CS-. In contrast, the AD group showed a marked impairment in conditioning, failing to exhibit significant conditioned fear responses. This failure to acquire conditioned responses could not be attributed to diminished responding by patients, relative to controls, to the aversive US. The results indicate that fear conditioning, an amygdala-dependent form of memory, is impaired in AD. These findings complement previous reports of impairments in declarative emotional memory in AD by demonstrating that a basic form of nondeclarative emotional memory is also impaired in AD.     

Hippocampal lesions disrupt classical conditioning of cross-modality reversal learning of the rabbit nictitating membrane response

The role of the hippocampus and subiculum in classical conditioning of tone-light discrimination reversal learning of the rabbit nictitating membrane response was investigated using aspiration lesions of both limbic structures. Only two of seven animals with hippocampal-subicular damage successfully reached reversal criteria within 21 days of conditioning, although all hippocampectomized animals learned the initial discrimination at rates equivalent to those of two control groups. Thus, previously reported deficits in two-tone reversal learning seen after similar lesions are not due to increased within-modality generalization to the conditioned stimuli (CS) serving as the CS+ and CS-.     

Retention of CS-US association learned under ketamine anesthesia

In order to assess learning possibilities under ketamine anesthesia, rats underwent a pavlovian conditioning (tone paired with foot-shock), and were subsequently tested in a conditioned suppression of lever pressing for food paradigm. First, two sessions of 70 trials were conducted, on 4 groups of anesthetized rats: a conditioned group (group C) experienced the tone (CS+) always paired with foot-shocks, whereas a discriminative group (group D) was submitted to a differential conditioning (CS+ followed by the US and a CS- always presented alone). Another group (habituated, H) received presentations of tone (CS+) alone and the fourth group (naive animals, N) was never anesthetized. Seven days later, while rats were pressing the lever to get food, CS+ presentations elicited a significant conditioned suppression of instrumental responses only in groups C and D. Moreover, a differential effect between CS+ and CS- was obtained for group D. These results indicate that a Pavlovian conditioning can occur under anesthesia, and then that its memory can last at least 7 days. The particular specific effects of ketamine are discussed to account for these results.     

An electrophysiological characterization of ventral tegmental area dopaminergic neurons during differential pavlovian fear conditioning in the awake rabbit.

Recent research has suggested that the mesencephalic dopaminergic (DA) system is activated by stress. For example, alterations in DA metabolites have been found in the ventral tegmental area (VTA) following footshock and immobilization in the rat [15,37]. Furthermore, this activation appears selective to DA neurons within the VTA since no changes were observed within the substantia nigra [15,16]. While this research suggests that DA neurons in the VTA are activated by aversive events, there has been a paucity of electrophysiological research designed to examine the sensory response characteristics of these DA neurons, and in particular their response to stimuli which predict aversive events. The present study was conducted to investigate the response characteristics of DA neurons within the VTA of the awake rabbit to acoustic stimuli which, via Pavlovian aversive conditioning procedures, came to predict the occurrence of a mild shock to the pinna. 45%, of the neurons meeting pre-established criteria for DA neurons demonstrated either significant excitation or inhibition to conditioned aversive stimuli. These neurons responded differentially to CS+ and CS- presentations. Some of these neurons (65%) demonstrated a greater increase in activity during the CS+ compared to the CS-, some (22%,) demonstrated a greater decrease in activity during the CS+ compared to the CS- and some (13%) demonstrated a greater increase in activity during the CS- compared to the CS+. Further, conditioned heart rate responses in the rabbits occurred during the recording of a majority of these neurons. These overall results suggest that conditioned aversive stimuli can affect the firing of VTA DA neurons and that these neurons comprise a heterogenous population with respect to their response profiles.     

Neural correlates of subjective CS/UCS association in appetitive conditioning

Explicit knowledge of conditioned stimulus (CS)/unconditioned stimulus (UCS) associations is proposed as important factor in classical conditioning. However, while previous studies have focused on its roles in fear conditioning, it has been neglected in the context of appetitive conditioning. The present functional magnetic resonance study aimed to investigate neural activation and functional connectivity linked to subjective CS/UCS association in appetitive conditioning. In total, 85 subjects participated in an appetitive acquisition procedure in which a neutral stimulus (CS+) was paired with a monetary reward, while another neutral stimulus (CS‐) was never paired with the reward. Directly afterwards, subjective CS/UCS association was assessed by measuring the extent to which the CS+ was thought to be associated with the UCS compared to the CS‐. Close relationships were established between subjective CS/UCS association and activations in the primary visual cortex (V1) during the early phase of conditioning and in the striatum during the late conditioning phase. In addition, we observed inverse relationships between subjective CS/UCS association and both V1/ventromedial prefrontal cortex (vmPFC) and striatal/vmPFC connectivity. The results suggest the involvement of decoupling vmPFC connectivity in reward learning in general and the roles of attentional processes in the formation of the subjective CS/UCS association during the early phase and reward prediction during the late phase of appetitive conditioning.     

Methods for developmental studies of fear conditioning circuitry.

Psychophysiologic studies use air puff as an aversive stimulus to document abnormal fear conditioning in children of parents with anxiety disorders. This study used functional magnetic resonance imaging (fMRI) to examine changes in amygdala activity during air-puff conditioning among adults. Blood oxygen level-dependent (BOLD) signal was monitored in seven adults during 16 alternating presentations of two different colored lights (CS+ vs. CS-), one of which was consistently paired with an aversive air puff. A region-of-interest analysis demonstrated differential change in BOLD signal in the right but not left amygdala across CS+ versus CS- viewing. The amygdala is engaged by pairing of a light with an air puff. Given that prior studies relate air-puff conditioning to risk for anxiety in children, these methods may provide an avenue for directly studying the developmental neurobiology of fear conditioning.     

Fear conditioning in adolescents with anxiety disorders: results from a novel experimental paradigm

OBJECTIVE: Considerable research examines fear conditioning in adult anxiety disorders but few studies examine youths. Adult data suggest that anxiety disorders involve elevated fear but intact differential conditioning. We used a novel paradigm to assess fear conditioning in pediatric anxiety patients. METHOD: Sixteen individuals with anxiety disorders and 38 healthy comparisons viewed two photographs of actresses displaying neutral expressions. One picture served as the conditioned stimulus (CS), paired with a fearful expression and a shrieking scream (CS+), whereas the other picture served as a CS unpaired with the aversive outcome (CS-). Conditioning was indexed by self-reported fear. Subjects participated in two visits involving conditioning and extinction trials. RESULTS: Both groups developed greater fear of the CS+ relative to CS-. Higher fear levels collapsed across each CS characterized anxious relative to healthy subjects, but no significant interaction between group and stimulus type emerged. Fear levels at visit 1 predicted avoidance of visit 2. Fear levels to both CS types showed stability even after extinction. CONCLUSIONS: Consistent with adult data, pediatric anxiety involves higher fear levels following conditioning but not greater differential conditioning. Extending these methods to neuroimaging studies may elucidate neural correlates of fear conditioning. Implications for exposure therapies are discussed.