Search

Reinforcement learning capitalizes on prediction errors (PEs), representing the deviation of received outcomes from expected outcomes. Mediofrontal event-related potentials (ERPs), in particular the feedback-related negativity (FRN)/reward positivity (RewP), are related to PE signaling, but there is disagreement as to whether the FRN/RewP encode signed or unsigned PEs. PE encoding can potentially be dissected by time-frequency analysis, as frontal theta [4–8 Hz] might primarily encode poor outcomes, while central delta [1–3 Hz] might instead represent rewarding outcomes. However, cortical PE signaling in negative reinforcement is still poorly understood, and the role of cortical PE representations in behavioral reinforcement learning following negative reinforcement is relatively unexplored. We recorded EEG while participants completed a task with matched positive and negative reinforcement outcome modalities, with parametrically manipulated single-trial outcomes producing positive and negative PEs. By analyzing response times (RTs) following reinforcing outcomes, we first demonstrated that PEs systematically influence future behavior in both positive and negative reinforcement conditions. Then, we mapped PEs onto whole scalp ERP and time-frequency (theta & delta) activity separately for positive and negative reinforcement. In negative reinforcement conditions, mediofrontal ERPs positively signaled unsigned PEs in a time window encompassing the P2 potential, and negatively signaled signed PEs for an extended time window encompassing the FRN/RewP and frontal P3 (an “aversion positivity”). Time-frequency decomposition demonstrated that central delta power increased parametrically with increasingly aversive outcomes, contributing to the “aversion positivity”. Finally, negative reinforcement ERPs correlated with RTs on the following trial, suggesting cortical PEs guide behavioral adaptations. Positive reinforcement PEs did not influence ERP or time-frequency activity, despite significant behavioral effects. These results demonstrate that mediofrontal PE signals are a mechanism underlying negative reinforcement learning, and that delta-band power increases for aversive outcomes might contribute to the “aversion positivity.”

The reward positivity (RewP) is a putative biomarker of depression. Careful control of stimulus properties and manipulation of both stimulus valence and salience could facilitate interpretation of the RewP. RewP interpretation could further be improved by investigating functional outcomes of a blunted RewP in depression, such as reduced memory for rewarding outcomes. This study sought to advance RewP interpretation first by advancing task design through use of neutral (i.e., draw) control trials and counterbalanced feedback stimuli. Second, we examined the RewP's association with memory and the impact of depression. Undergraduates completed self-report measures of depression and anhedonia prior to a modified doors task in which words were displayed in colored fonts that indicated win, loss, or draw feedback. Memory of the feedback associated with each word (i.e., source memory) was tested. Results showed that RewP response to wins was more positive than to losses, which was more positive than to draws. The RewP was not associated with depression or anhedonia. The low depression group showed a source memory advantage for win words, but the high depression group did not. Source memory showed small relations to the RewP, but these did not survive Bonferroni correction. Results suggest the RewP is sensitive to salience and highlight challenges in detecting an association between the RewP and depression in modified doors tasks. Findings indicate that depression is related to dysfunctional source memory for reward but not loss and that future research should probe the possible associations between the RewP and memory in depression.

BackgroundCannabis Use Disorder (CUD) has been linked to environmental, personality, mental health, neurocognitive and neurobiological risk factors. While many studies have revealed gender differences in CUD, the relative importance of these complex factors by gender has not been described.MethodsWe conducted a data-driven examination of gender differences in CUD in a community sample of young adults (Human Connectome Project [HCP]; n = 1204, 54% female). We employed state-of-the-art machine learning methods [gradient tree boosting, XGBoost] in combination with novel factor ranking tools [SHapley’s Additive exPlanations (SHAP)] as an ‘explainable machine learning approach’ in the multimodal data collected by the HCP (phenotypic and brain data).ResultsWe were able to successfully classify both cannabis dependence and cannabis use levels. Previously identified environmental, personality, mental health, neurocognitive, and brain factors highly contributed to the classification. Predominantly-male risk factors included personality (high openness), mental health (high externalizing, high childhood conduct disorder, high fear somaticism), neurocognitive (impulsive delay discounting, slow working memory performance) and brain (low hippocampal volume) factors. Conversely, predominantly-female risk factors included environmental (low education level, low instrumental support) factors.ConclusionsOur data-driven analysis of gender differences in the multimodal risk factors underlying cannabis dependence and use levels demonstrate that environmental factors contribute more strongly to CUD in women, whereas individual factors such as personality, mental health and neurocognitive factors have a larger importance in men. This warrants further investigations, and suggests the importance of understanding how these differences relate to the development of effective treatment approaches.

Negative Urgency (NU) is a prominent risk factor for hazardous alcohol use. While research has helped elucidate how NU relates to neurobiological functioning with respect to alcohol use, no known work has contextualized such functioning within existing neurobiological theories in addiction. Therefore, we elucidated mechanisms contributing to the NU–hazardous alcohol use relationship by combining NU theories with neurobiological dual models of addiction, which posit addiction is related to cognitive control and reinforcement processing. Fifty-five undergraduates self-reported NU and hazardous alcohol use. We recorded EEG while participants performed a reinforced flanker task. We measured cognitive control using N2 activation time-locked to the incongruent flanker stimulus, and we measured reinforcement processing using the feedback-related negativity (FRN) time-locked to better-than-expected negative reinforcement feedback. We modeled hazardous drinking using hierarchical regression, with NU, N2, and FRN plus their interactions as predictors. The regression model significantly predicted hazardous alcohol use, and the three-way interaction (NU × N2 × FRN) significantly improved model fit. In the context of inefficient processing (i.e., larger N2s and FRNs), NU demonstrated a strong relationship with hazardous alcohol use. In the context of efficient processing (i.e., smaller N2s and FRNs), NU was unrelated to hazardous alcohol use. Control analyses ruled out the potential impact of other impulsivity subscales, individual differences in dimensional negative affect or anxiety, and use of substances other than alcohol, and post hoc specificity analyses showed that this effect was driven primarily by heavy drinking, rather than frequency of drinking. This analysis provides preliminary evidence that brain mechanisms of cognitive control and reinforcement processing influence the relationship between NU and hazardous alcohol use, and confirms a specific influence of negative reinforcement processing. Future clinical research could leverage these neurobiological moderators for substance misuse treatment.

Cognitive control processes encompass many distinct components, including response inhibition (stopping a prepotent response), proactive control (using prior information to enact control), reactive control (last-minute changing of a prepotent response), and conflict monitoring (choosing between two competing responses). While frontal midline theta activity is theorized to be a general marker of the need for cognitive control, a stringent test of this hypothesis would require a quantitative, within-subject comparison of the neural activation patterns indexing many different cognitive control strategies, an experiment lacking in the current literature. We recorded EEG from 176 participants as they performed tasks that tested inhibitory control (Go/Nogo Task), proactive and reactive control (AX-Continuous Performance Task), and resolving response conflict (Global/Local Task-modified Flanker Task). As activity in the theta (4–8 Hz) frequency band is thought to be a common signature of cognitive control, we assessed frontal midline theta activation underlying each cognitive control strategy. In all strategies, we found higher frontal midline theta power for trials that required more cognitive control (target conditions) versus control conditions. Additionally, reactive control and inhibitory control had higher theta power than proactive control and response conflict, and proactive control had higher theta power than response conflict. Using decoding analyses, we were able to successfully decode control from target trials using classifiers trained exclusively on each of the other strategies, thus firmly demonstrating that theta representations of cognitive control generalize across multiple cognitive control strategies. Our results confirm that frontal midline theta-band activity is a common mechanism for initiating and executing cognitive control, but theta power also differentiates between cognitive control mechanisms. As theta activation reliably differs depending on the cognitive control strategy employed, future work will need to focus on the differential role of theta in differing cognitive control strategies.

Alcohol use disorder (AUD) has high prevalence and adverse societal impacts, but our understanding of the factors driving AUD is hampered by a lack of studies that describe the complex neurobehavioral mechanisms driving AUD. We analyzed causal pathways to AUD severity using Causal Discovery Analysis (CDA) with data from the Human Connectome Project (HCP; n = 926 [54% female], 22% AUD [37% female]). We applied exploratory factor analysis to parse the wide HCP phenotypic space (100 measures) into 18 underlying domains, and we assessed functional connectivity within 12 resting-state brain networks. We then employed data-driven CDA to generate a causal model relating phenotypic factors, fMRI network connectivity, and AUD symptom severity, which highlighted a limited set of causes of AUD. The model proposed a hierarchy with causal influence propagating from brain connectivity to cognition (fluid/crystalized cognition, language/math ability, & working memory) to social (agreeableness/social support) to affective/psychiatric function (negative affect, low conscientiousness/attention, externalizing symptoms) and ultimately AUD severity. Our data-driven model confirmed hypothesized influences of cognitive and affective factors on AUD, while underscoring that addiction models need to be expanded to highlight the importance of social factors, amongst others., Rawls and colleagues use an advanced statistical approach to identify causal neurobehavioral mechanisms underlying Alcohol Use Disorder. Their findings support current multifactorial models of addiction, but also highlight the importance of social factors in addiction maintenance.

We demonstrate a data-driven approach for calculating a “causal connectome” of directed connectivity from resting-state fMRI data using a greedy adjacency search and pairwise non-Gaussian edge orientations. We used this approach to construct n=442 causal connectomes. These connectomes were very sparse in comparison to typical Pearson correlation-based graphs (roughly 2.25% edge density) yet were fully connected in nearly all cases. Prominent highly connected hubs of the causal connectome were situated in attentional (dorsal attention) and executive (frontoparietal and cingulo-opercular) networks. These hub networks had distinctly different connectivity profiles: attentional networks shared incoming connections with sensory regions and outgoing connections with higher cognitive networks, while executive networks primarily connected to other higher cognitive networks and had a high degree of bidirected connectivity. Virtual lesion analyses accentuated these findings, demonstrating that attentional and executive hub networks are points of critical vulnerability in the human causal connectome. These data highlight the central role of attention and executive control networks in the human cortical connectome and set the stage for future applications of data-driven causal connectivity analysis in psychiatry.

Aggression and violence are social behaviors that exact a significant toll on human societies. Individuals with aggressive tendencies display deficits in effortful control, particularly in affectively charged situations. However, not all individuals with poor effortful control are aggressive. This study uses event-related potentials (ERPs) recorded from a large sample (n = 75 undergraduates) to decompose the chronology of neural mechanisms underlying the ability to effortfully-control behavior, and then explores whether deficits in these cognitive functions might then lead to aggressive behavior. This study investigated which ERPs moderate the effortful control - aggression association. We examined three successive ERP components, the P2, N2, and P3, which have been associated with attentional orienting, response conflict, and working memory updating, for stimuli that required effortful control. N2 amplitudes were larger for trials requiring a switch from a preplanned action strategy than trials where a preplanned action strategy was followed. Furthermore, results indicated that N2 activation, but not P2 or P3 activation, moderated the relationship between effortful control and aggression. Our results suggest that small (less negative) N2s moderate the association between effortful control and aggression. These effects were present only in negative contexts, and only for high-conflict trials. Results suggest that individual differences in neural processing efficiency contributes to the execution of effortfully controlled behavior and avoidance of aggression.

A number of studies have indicated that violent video gameplay is associated with higher levels of aggression and that desensitization and selective attention to violent content may contribute to this association. Utilizing an emotionally-charged rapid serial visual presentation (RSVP) task, the current study used two event-related potentials (ERPs) - the N1 and P3 - that have been associated with selective attention and desensitization as neurocognitive mechanisms potentially underlying the connection between gameplay and higher levels of aggression. Results indicated that video game players and non-players differed in N1 and P3 activation when engaged with emotionally-charged imagery. Additionally, P3 amplitudes moderated the association between video gameplay and aggression, indicating that players who display small P3 amplitudes also showed heightened levels of aggression. Follow-up moderational analyses revealed that individuals who play games for many hours and show more negative N1 amplitudes show smaller P3 activation. Together, our results suggest that selective attention to violent content and desensitization both play key roles in the association between video gameplay and aggression.

Theories of emotion-cognition interactions suggest that emotional valence can both facilitate or limit cognitive performance. One cause for the mixed findings may be the order (random versus non-random presentation) in which emotional stimuli are presented. To investigate the impact of stimuli order on cognitive control processing, EEG data were recorded as 130 undergraduate students (M age = 22.2, SD = 5.4; 79 female) completed a modified version of the AX-Continuous Performance Task in which the cue was followed by an emotionally-valenced image (positive, negative, and neutral). Specifically, the task was designed so that valenced images were presented in either a block or random order, prior to probe presentation. We examined two event-related potentials (ERPs), the N2, which reflects aspects of cognitive control, and the late positive potential (LPP), which reflects attention allocation to emotional stimuli. We assessed the impact of emotionally oriented attention (LPP) on downstream cognitive control (N2) and how this relationship might differ for a block versus random (order of emotional image) task design. Consistent with the LPP literature, we found a main effect of image valence with the negative trials showing larger LPPs than the positive and neutral trials. For N2s, we found that the negative trials were associated with smaller N2s than both the positive and neutral trials. We observed that as LPP amplitude increased, subsequent N2 amplitude was reduced, specifically for negative trials in the random design. These results suggest an emotion-related depletion of neural cognitive resources. Lastly, we found larger N2s for the block design versus the random design. Together, these results indicate the importance of paying attention to both trial order (block versus random) and within trial stimulus sequence when designing emotion induction tasks.

Prediction errors (PEs) encode representations of rewarding and aversive experiences and are critical to reinforcement processing. The feedback-related negativity (FRN), a component of the event-related potential (ERP) that is sensitive to valenced feedback, is believed to reflect PE signals. Reinforcement is also studied using frontal midline theta (FMΘ) activity, which peaks around the same time as the FRN and increases in response to unexpected events compared to expected events. We recorded EEG while participants completed a monetary incentive delay (MID) task that included positive reinforcement and negative reinforcement conditions with multiple levels of the outcome, as well as control conditions that had no reinforcement value. Despite the overlap of FRN and FMΘ, these measures indexed dissociable cognitive processing. The FRN was sensitive to errors in both positive and negative reinforcement but not in control conditions, while frontal theta instead was sensitive to outcomes in positive reinforcement and control conditions, but not in negative reinforcement conditions. The FRN was sensitive to the point level of feedback in both positive and negative reinforcement, while FMΘ was not influenced by the feedback point level. Results are consistent with recent results indicating that the FRN is influenced by unsigned PEs (i.e., a salience signal). In contrast, we suggest that our findings for frontal theta are consistent with hypotheses suggesting that the neural generators of FMΘ are sensitive to both negative cues and the need for control.

When multiple competing responses are activated, we respond more slowly than if only one response is activated (response conflict). Conflict-induced slowing is reduced for consecutive high-conflict stimuli, an effect known as conflict adaptation. Verguts and Notebaert’s (2009) adaptation by binding theory suggests this is due to Hebbian learning of cognitive control, potentiated by the response of the locus coeruleus norepinephrine (NE) system. Phasic activity of the NE system can potentially be measured non-invasively in humans by recording the P3 component of the event-related potential (ERP), and the P3 is sensitive to conflict adaptation. Bouret and Sara’s (2005) network reset theory suggests that phasic NE might functionally reset ongoing large-scale network activity, generating synchronous neural population activity like the P3. To examine the possibility that network reset contributes to conflict effects in the P3, we recorded high-density EEG data while subjects performed a flanker task. As expected, conflict and conflict adaptation modulated P3 amplitudes. Brain-behavior correlation analyses indicated that activity during the rise of the P3 was related to RT and predicted RT differences due to conflict. More importantly, phase of delta oscillations not only predicted reaction time differences between low-conflict and high-conflict conditions, but delta phase reset also predicted the amplitude of the P3. Delta oscillations exhibited dominant peaks both pre and post-stimulus, and delta at stimulus onset predicted the post-stimulus ERP, in particular the N2 and P3. This result bridges human EEG with basic mechanisms suggested by computational neural models and invasive patient recordings, namely that salient cognitive events might reset ongoing oscillations leading to the generation of the phase-locked evoked potential. We conclude that partial phase reset is a cortical mechanism involved in monitoring the environment for unexpected events, and this response contributes to conflict effects in the ERP. These results are in line with theories that phasic NE release might reset ongoing cortical activity, leading to the generation of ERP components like the P3.

Emotional eating is an attempt to avoid, control, or cope with negative emotions through eating a large amount of calorie dense sweet and/or high fat foods. Several factors, including various attentional mechanisms, negative affect, and stress, impact emotional eating behavior. For example, attentional narrowing on negative events may increase attentional stickiness and thereby prevent the processing of more peripheral events, such as eating behavior. This study contributes to the extant literature by examining the neural correlates underlying the attentional conflict between processing negative events and regulating behavior within a task that emulates how negative life experiences might contribute to unrestrained eating behavior. We explore this question within a normative sample that varies in their self-reported anxiety symptoms. Dense-array EEG was collected while participants played the attentional blink game—a task in which excessive attentional resource allocated to one event (e.g., negative picture) interferes with the adequate attentional processing of a second event that requires action. To assess the attentional conflict, we measured N2 activation, an event-related potentials (ERPs; averaged EEG) associated with conflict processing. Results revealed that N2 activation moderates the association between anxiety and emotional-eating behavior. Thus, increased anxiety combined with more negative N2 activation can contribute to emotional-eating behavior. These results are discussed in the context of ineffective conflict processing contributing to poor emotion regulation.