Aversive learning in the visual brain: Generalization across feature dimensions
University of Florida
Processing capabilities for many low-level visual features are experientially malleable, aiding sighted organisms in adapting to dynamic environments. In this presentation, we discuss how visuocortical responses change as human observers learned to associate exemplars drawn from a given feature dimension with aversive outcomes. Using classical aversive conditioning together while recording dense-array EEG and pupillometry, we tested the pre-registered hypotheses of either sharpening or generalization for a range of feature dimensions, including orientation, motion direction, object category, and spatial location. Models of gaussian (generalization) or difference-of-gaussian (sharpening) changes after, compared to pre-conditioning were directly compared in a Bayesian framework. We found that visuocortical responses were selectively heightened when viewing aversively paired features for all feature dimensions. In the case of orientation, motion direction, and spatial location, effects displayed a non-linear, difference-of-gaussian profile across neighboring exemplars on a feature gradient, consistent with suppressive surround modulation of non-prioritized features. Measures of alpha band (8 – 12.8 Hz) activity and pupil diameter showed evidence of generalization. These results indicate that aversive conditioning of low-level visual prompts sharpened tuning in visual cortex. By contrast, aversive conditioning of higher-level features such as object category prompts linearly graded (generalization) modulation in visual cortex. These effects mirror the effects seen for top-down influences indexed by alpha power reduction and autonomic responses, also showing generalization. We summarize these changes in a simple computational model of adaptive population tuning as a function of experience.
Externally induced cross-regional synchronisation modulates human associative memory: empirical and computational evidence
1Institute for Neuroscience and Psychology, University of Glasgow, United Kingdom; 2School of Psychology, University of Birmingham, United Kingdom
Neural synchronisation between sensory brain regions has been suggested as a mechanism to bind multisensory information into long-term memory. Using a multisensory entrainment paradigm, we demonstrated a causal role of theta synchronisation between sensory regions in enhancing human associative memory. EEG data reveals that this externally induced theta phase synchrony supports associative memory on a trial-by-trial basis. Furthermore, a computational model that implements two key hippocampal learning mechanisms found in animal studies, spike-timing-dependent-plasticity (STDP) and theta-phase-dependent learning, successfully simulates our findings. In the model, hippocampal weight changes depend on the timing between spikes of pre-synaptic and post-synaptic neurons, which is also modulated by the phase of an ongoing theta oscillation. Stretching the temporal window for plasticity in our simulated learning rule fails to corroborate the empirical data, emphasising the importance of neuronal frequency working in tandem with plasticity windows (~25 ms). To experimentally test this hypothesis, we modified our paradigm by changing the modulation frequency to gamma. Indeed, participants’ recall accuracy was best in the 0° phase offset condition and was significantly better than the 270° condition, which has the longest delay between sensory stimuli. These results can be seen as direct support for the role of STDP in human multisensory memory association, which restricts associative binding largely to a narrow time window. In conclusion, our findings provide causal evidence for neural synchronisation in human memory. The empirical evidence, together with the computational evidence, links circuit-level mechanisms described in the animal literature with data from human associative memory studies.
4 Hz phase-synchronized stimulus presentation improves contingency knowledge and affective evaluation in a fear conditioning task
1Universität Osnabrück, Deutschland; 2University of Florida, USA
Most memory traces combine information from multisensory input. Animal studies show that theta-synchronized neuronal activity (4-8 Hz) binds multimodal associations. Recent human data revealed improved declarative memory for video-tone pairings when the input was presented in theta-phase synchronization. Since classical fear conditioning, a non-declarative, emotional memory paradigm, is mostly based on associating a conditioned stimulus (CS) in one modality with an aversive unconditioned stimulus (US) in another modality, the present study examines the effects of theta-band stimulus synchronization on fear memory formation in humans.
Using a fear generalization procedure, one of five visual gratings (that only varied in orientation) was paired with an aversive auditory US, and thus served as the CS+. Both, luminance of the visual CS (one CS+, four CS-) and the volume of the auditory US were modulated at 4 Hz. To manipulate the synchrony between visual and auditory input during fear acquisition, one group (N = 20) received 12 trials of synchronous CS-US pairing, whereas the control group (N = 20) received the 12 CS-US pairs out-of-phase (90°, 180°, and 270° shift).
Phase-synchronized input improved the CS-US contingency knowledge and the subjective valence and arousal ratings, leading to a narrower generalization around the CS+ compared to out-of-phase input. Physiological arousal (skin conductance) and visuocortical engagement (steady-state visually evoked potentials, ssVEP) were unaffected by the synchronization manipulation, although both measures were increased for the CS+. Our data suggest that theta-phase synchronization aids declarative aspects of fear memory formation, such as learned affective evaluation and contingency knowledge.
Prefrontal oscillations and the immediate extinction deficit in healthy men
Department of Psychology, University of Osnabrueck, Osnabrueck, Germany
The interval between fear acquisition and extinction learning is thought to influence extinction recall. In most fear-conditioning studies, short intervals (e.g. 10 min) were shown to impair extinction recall compared to long intervals (e.g., 24 hours), while fear recall was unaffected. This phenomenon is known as Immediate Extinction Deficit (IED). Rodent studies suggest that stress-induced inhibition of the medial prefrontal cortex (mPFC) during immediate extinction learning causes the IED. However, studies examining the neural correlates of the IED in humans are missing. We used a well-established fear-conditioning paradigm covering habituation, fear acquisition, extinction learning, and recall. 40 healthy men were randomly assigned to extinction learning either 10 min after fear acquisition (immediate extinction learning) or 24h afterwards (delayed extinction learning). Four pictures of male faces were used as conditioned stimuli (CS). During fear acquisition, two CS (CS+) were paired with an aversive unconditioned stimulus (US), whereas the other two were not (CS-). During extinction learning, only one CS+/CS- pair was presented. Fear recall (responses to the not-extinguished CS+N and CS-N) and extinction recall (responses to the previously extinguished CS+E vs. CS-E) were assessed 24h after extinction learning. EEG oscillations in the theta and gamma range were source-localized during all learning phases. On the peripheral level CS-related skin conductance responses (SCRs) and cardiac responses were assessed. First SCR results show the expected better extinction recall in the delayed extinction group. EEG and cardiac-response analysis are still under progress and will be presented.
Noradrenergic modulations of fear conditioning: Yohimbine potentiates fear-conditioned bradycardia, N170, and late positive potential amplitudes
1Department of Psychology, University of Marburg, Germany; 2Department of Psychology, University of Giessen, Germany; 3Department of Psychiatry, Harvard Medical School, USA; 4Department of Psychology, University of Vienna, Austria
Fear conditioning is an important model for understanding the etiology and maintenance of anxiety disorders, while fear extinction is considered to reflect a central learning mechanism underlying exposure therapies. Hyperconsolidation of aversive conditioned associations and poor extinction consolidation have been hypothesized to be crucial in the acquisition of pathological fear. Previous animal and human research has pointed to a potential role of catecholaminergic neurotransmitters, particularly noradrenaline and dopamine, in acquiring emotional memories. Here, we investigated whether the noradrenergic alpha-2 adrenoreceptor antagonist yohimbine and the dopaminergic D2 receptor antagonist sulpiride modulate human consolidation of conditioned and extinguished fear conditioning and extinction. Fifty-four participants received yohimbine (10 mg, n = 18), sulpiride (200 mg, n = 18), or placebo (n = 18) between fear acquisition and extinction. The yohimbine group showed increased alpha-amylase activity, confirming a successful manipulation of central noradrenergic release. We assessed recall of conditioned (non-extinguished CS+ vs. CS-) and extinguished fear (extinguished CS+ vs. CS-) 24 hours later. Importantly, potentiated fear bradycardia and larger amplitudes of the N170 and LPP ERP components indicated that noradrenergic yohimbine treatment (compared to placebo and sulpiride) enhanced fear recall on day 2. These results suggest that yohimbine potentiated cardiac and central electrophysiological signatures of fear memory consolidation. In conclusion, our findings elucidate the key role of noradrenaline in strengthening conditioned fear.