Individual talks: Learning

July 11th A2.09

The role of perceptual and conceptual properties in affective habituation

Oksana Itkes & Assaf Kron

Habituation is perhaps the most pervasive and evolutionary ancient form of learning, defined as attenuation of response following repeated exposure to a stimulus. The ability to habituate to affective information is especially important, as constant activation of a strong emotional response can be maladaptive in most everyday situations. Surprisingly, very little is known about the mechanism that underlies affective habituation. In a set of two experiments, we investigated the level of stimulus abstraction at which habituation of emotional response occurs. Specifically, we asked whether in the process of repeated exposure, affective habituation occurs for perceptual, conceptual and/or affective properties of the stimulus. To investigate this question, participants were repeatedly presented with an affective image, followed by a set of test images that shared perceptual, conceptual, or affective properties with the repeated stimulus, allowing us to compare the degree to which habituation can be generalized across different levels of stimulus abstraction. Results demonstrated that habituation across different components of the emotional response (self-reported feelings, facial expressions) were generalized up to the conceptual level of the repeated stimulus. These findings suggest that the conceptual system plays a role in affective learning.

Support for the goal-directedness of early emotional action tendencies from a TMS study

Maja Fischer, Chiara Fini, Marcel Brass & Agnes Moors

Traditional emotion theories suggest that emotional action tendencies are initially impulsive but might turn into instrumental behavior during concrete action planning or emotion regulation. Impulsive behaviors are assumed to stem from a rigid stimulus-driven process in which stimulus features are associated with a fixed action tendency. Instrumental behaviors stem from a flexible goal-directed process, in which the expected utilities of different behavior options are assessed and the behavior option with the highest expected utility is chosen. In line with a recent alternative goal-directed theory of emotion (Moors, 2017; Moors, Boddez, & De Houwer, 2017), we hypothesized that initially impulsive emotional action tendencies can already be determined by a goal-directed process and can override any stimulus-driven process that might run in parallel. We conducted an approach-avoidance task in which 67 participants were randomly assigned either to a condition in which they merely observed positive and negative stimuli or a condition in which they were rewarded to approach negative stimuli and avoid positive stimuli. Action tendencies were measured via motor evoked potentials (MEP) after transcranial magnetic stimulation (TMS). Results showed that positive/negative stimuli led to stronger approach/avoidance tendencies in the observation condition, suggesting the operation of a stimulus-driven process linking positive/negative stimuli to approach/avoidance tendencies. In the response condition, we observed the opposite pattern in which positive/negative stimuli led to stronger avoidance/approach tendencies, suggesting that a goal-directed process was able to override the stimulus-driven process already at an early stage. This suggests that also early emotional action tendencies can be instrumental.

Comparing Three Models of Arousal in the Human Brain

Hadeel Haj Ali, Assaf Kron & Adam Anderson

Valence and arousal are frequently used to model the conscious experience of emotion. In this work, we distinguish between three versions of the valence-arousal model, according to how they interpret the arousal dimension. The first model assumes that arousal is separate qualia from bipolar valence. The second model interprets arousal as the intensity of bipolar valence, and the third model suggests that arousal is a linear combination of two separate unipolar dimensions of pleasant and unpleasant. Thirty participants viewed emotional pictures in the MRI scanner, while providing reports about their emotional response. Half of the reports were given with bipolar valence and arousal scales, and the other half with two unipolar scales for pleasant and unpleasant feelings. Using parametric modulation approach, we compared the three models in their ability to predict neural activation in arousal-related regions. The model "arousal as separate qualia" was estimated by traditional arousal self-reports, "arousal as intensity" model was estimated by absolute values of the bipolar valence scale, and "arousal as a linear combination of pleasant and unpleasant" was estimated by summation of unipolar pleasant and unpleasant scales. The overall results showed an advantage for the sum of pleasant and unpleasant over absolute values of valence, and over arousal, in predicting neural activation in arousal-related regions. No significant difference was found between arousal and absolute values of valence. These findings do not support arousal as distinct qualia from valence in human conscious experience.

Stress-induced Pavlovian bias in reward seeking behavior

Eva R. Pool, Alessio Giarrizzo, Yoann Stussi & David Sander

A common symptom across many clinical conditions such as binge eating, pathological gambling or drug addiction is the willingness to go to extraordinary lengths in order to obtain an object of desire, even though once obtained the object is not experienced as pleasurable. What are the mechanisms that make the human brain vulnerable to situations where choice behavior is hijacked in the service of outcomes that are not valued by the individual? I will present a study where we investigated the impact of stress on the balance between the Pavlovian and the goal-directed control on reward seeking behaviors. We tested the hypothesis that stress amplifies the Pavlovian control of reward-seeking behaviors, while hindering higher-level goal-directed control. To this end, we combined a Pavlovian-instrumental conflict paradigm with a stress induction manipulation and measured anticipation behaviors of visual rewards by means of eye-tracking techniques. We additionally used computational modeling approaches to characterize the specific learning components affected by stress. Altogether, our data provide insights into how stress modulates the balance between the Pavlovian and the goal-direct control, a key mechanism that may underlie dysfunctional reward-seeking behaviors in humans.

Neural markers of emotion dysregulation in acute trauma survivors predict chronic PTSD

Christine Larson

Trauma exposure is extremely common and increases risk for a host of negative health outcomes, most notably posttraumatic stress disorder (PTSD). Unfortunately, despite some progress, determining which acutely traumatized individuals are at risk for chronic PTSD remains difficult. Furthermore, the neural mechanisms predicting longitudinal risk for PTSD are almost completely unexplored. To advance prediction of risk for chronic PTSD, we measured recruitment of the neural circuitry instantiating emotion regulation within two weeks of trauma exposure and the extent to which these neural markers predicted PTSD symptoms six months post-trauma. Dysregulation in both established PTSD networks (e.g., amygdala, ventromedial prefrontal cortex, anterior cingulate cortex), as well as less-studied regions, including the striatum and visual cortex predicted PTSD symptoms at six-month follow-up. Of note, there was some specificity of networks involved in avoidance, hyperarousal, and re-experiencing symptoms, which highlights the need to consider the symptom clusters separately, rather than in aggregate as is the norm in the imaging literature on chronic PTSD. More broadly, these results point to specific mechanisms underlying acute post-trauma emotion dysregulation that can be targeted to better identify trauma survivors at risk for chronic PTSD, and potentially to optimize early interventions.