Posters

A network analysis of motivational and affective individual differences and error monitoring

Anna Grabowska  1, 2, Filip Sondej  2  , Magdalena Senderecka  2  

1 : Doctoral School in the Social Sciences, Jagiellonian University

2 : Institute of Philosophy, Jagiellonian University

Error monitoring is influenced by a complex interplay of motivational and affective factors. Numerous studies investigating these associations have yielded inconsistent results, which may be due to differences in symptoms studied and covariates accounted for. To address these inconsistencies and methodological gaps, we utilized network analysis.. We employed four regularized Gaussian Graphical Models on a non-clinical population (N=260, aged 18-40) to examine the conditional dependence between the amplitude of response-related potentials (error-related negativity, ERN; correct-related negativity, CRN) and 29 self-report measures related to anxiety, depression, obsessive thoughts, compulsive behavior, reward, and punishment, while adjusting for covariates: age, handedness, and latency of ERN and CRN. We then validated our results on an independent sample of 112 healthy participants. ERN amplitude was uniquely associated with rumination, avoidance of ambiguity, thought suppression, reward sensitivity, and guilt sensitivity. Upon covariate adjustment, ERN remained uniquely linked to rumination, avoidance of ambiguity, and reward sensitivity. ERN was not directly associated with trait anxiety. Trait anxiety and behavioral inhibition were the most central nodes in all networks. These results underscore the nuanced effects of affective and motivational traits on error processing. Additionally, they suggest that some apparent links between ERN amplitude and individual differences may be spurious and potentially result from shared common causes.

Attention in neural networks for cognitive flexibility and stability 

Xiaoyu Zhang  1, Tom Verguts  1  , Mehdi Senoussi  2  

1 : Ghent University

2 : Université Toulouse II

CLLE-ERSS (UMR 5263)

Making your way in today's complex world requires two apparently contradictory abilities: One must flexibly adapt to novel tasks, but at the same time one must also retain the ability to solve earlier tasks. However, as tools to study such flexibility, conventional neural networks are notoriously plagued by severe problems. First, they require a lot of training for each new task (little transfer between tasks that share components); and second, they suffer from catastrophic inference, so they forget older information. To address these problems, we propose that attention can support neural networks. Attention is a key concept in psychology and neuroscience, and has recently also become influential in AI (in the context of transformers). Yet, the attention in transformers is cognitively implausible. Here, we seek to develop neural networks that learn to attend, but in a manner inspired by cognitive neuroscience. We investigate how the resulting model implements cognitive flexibility, and in particular generalization to (partially) novel tasks, and robustness in the face of potential interference. We fit the model to two well-established cognitive tasks: the modularity task and C-PRO task (Ito et al., 2022) , and compare it to traditional neural networks. Our findings reveal that neural networks equipped with multiplicative attentional gating exhibit significantly improved generalization and robustness compared to traditional networks. These insights hold meaningful implications for how attention is implemented in biological networks, and how it supports cognitive flexibility demonstrated by humans in adapting to various environments and tasks.

Decision Making in risky condition: a study on the impact of stress on decision-making 

Margaux Pourady  1, Alexandre Pouchot  1  , Andrea Desantis  1, 2  , Sebastien Scanella  3  , Stefania Ficarella  1  

1 : DTIS, ONERA [Salon-de-Provence] ONERA

2 : Institut de Neurosciences de la Timone (UMR 7289), CNRS, Aix-Marseille Université, Marseille

Institut de Neurosciences de la Timone

3 : DMIA, ISAE, Université de Toulouse, 10 Avenue E. Belin 31055 Toulouse Cedex 4

Institut Supérieur de l'Aéronautique et de l'Espace (ISAE)

Acute stress has been shown to significantly influence decision-making processes. For example, in high-pressure situations, such as those experienced by aircraft pilots, individuals may prioritize visual stimuli while disregarding important auditory alarms in the cockpit. This phenomenon suggests that stress can disrupt the balance between processing relevant and irrelevant information, ultimately affecting decision-making. To investigate this hypothesis, we designed a complex decision-making task. Participants are asked to take rapid decisions, based on visual or auditory information, regarding whether to land on a runway or initiate a go-around procedure. We manipulate the relevance of auditory stimuli: in one condition, decisions are based solely on auditory information, while in another, auditory signals are irrelevant, requiring participants to rely on visual cues only. Two groups of participants will complete the task. One group will undergo a stress induction phase prior to the landing task, while the other will undergo a control, stress-free induction. The stress induction phase will involve a modified version of the Mannheim Multicomponent Stress Test. We hypothesize that individuals with higher stress levels will allocate more attentional resources to irrelevant sounds, thereby neglecting relevant visual information. In contrast, we expect the control group to more efficiently ignore irrelevant auditory stimuli. This will be reflected by their behavioural performance. Notably, the stress group will make more errors and exhibit slower response times compared to the no-stress group. Additionally, we will record participants' electrical brain activity using electroencephalography (EEG) to gain insights into the processing of both relevant and irrelevant auditory information. We will measure ERP components, such as the parietal P300, associated with the depth of auditory information processing. We expect to observe similar P300 amplitudes for task-relevant and task-irrelevant sounds in the stress group, indicating difficulty in ignoring irrelevant information. Conversely, we anticipate a weaker P300 response to task-irrelevant sounds in the control group, reflecting their ability to efficiently filter out irrelevant cues. Importantly, we will assess participants' stress levels at various stages of the experiment using both objective (e.g., cortisol levels) and subjective measurements (e.g., via the STAI questionnaire). These data have not been collected yet, but testing is planned during the month of April/May.

Does an *illusion of conflict* trigger conflict adaptation? 

Anastassia Loukianov  1,  Aloïs Toussaint  1  , Axel Cleeremans  1  

1 : Consciousness Cognition & Computation Group, Université Libre de Bruxelles

To what extent does your subjective appraisal of a situation influence your cognition and behavior? This question which may appear trivial at first sight, becomes particularly relevant when your subjective experience misrepresents reality (e.g., in placebo or hypnotic suggestion). Here, we manipulate metacognition by inducing inaccurate beliefs about a situation through phenomenological suggestion (sham neurofeedback). To do so, we focused on the role of subjective experience in conflict tasks, which typically involve congruency relationships between prime and target stimuli. In such tasks, Desender et al. (2014) showed that sequential congruency effects (SCE - conflict adaptation) only occur when people actually experience the conflict. Here, taking inspiration from Palmeri et al. (1999), we would like to further explore what happens if people believe there is a conflict but there is no actual conflict (e.g., an illusory conflict). Are people going to adapt their behavior based on their (inaccurate) belief of conflict? How do these effects depend on people's metacognitive abilities? These kinds of effects are debated both in the neurofeedback literature and in the placebo literature. In this study, by providing participants with a sham neurofeedback (EEG) about the occurence of a conflict, we would like to investigate if this illusory experience of conflict can trigger conflict adaptation behavior (SCE) in a priming task similar to the one used by Desender et al. (2014).

Error analysis in visual word recognition among typical adult readers: a Conditional Accuracy Function analysis based on EMG 

Adrien Lorant  1  , Fanny Grisetto  1  , Gwendoline Mahé  1  , Clémence Roger  1  

1 : Laboratoire Sciences Cognitives et Sciences Affectives - UMR 9193

Université de Lille, Centre National de la Recherche Scientifique

Despite formal education, approximately 20% of the population experiences reading difficulties, leading to significant social repercussions. The investigation of the origin of errors during visual word recognition, inspired by research on cognitive control, enables to enhance the understanding of reading difficulties. This study aimed to investigate the sources of errors in a lexical decision task among 39 typical readers (33 women, 5 men, 1 non-binary) with an average age of 20.38 years, 22 of which are currently analysed. Participants were asked to perform a lexical decision task with four conditions — words with high and low frequency, transposed-letter pseudowords (e.g., cholocate for the word chocolate), and replaced-letter pseudowords (e.g., chofonate for the word chocolate). During the task, electromyographic activities (EMG) of the muscles involved in the responses were recorded in order to specify the performance in each trial (i.e. pure-correct, pure-error, and correct trials with a partial error). Conditional Accuracy Functions (CAFs) were performed to assess the accuracy of the first EMG response as a function of its latency, enabling to investigate the strength of the lexical capture for each condition.

Overall, the results confirm classical expectations, namely that the accuracy was higher for words with high than low frequency, and higher for replaced letter pseudowords than transposed-letter pseudowords. Interestingly, the preliminary CAFs analysis revealed that the first latency quantile differed significantly in the error rate between conditions. Specifically, an increased prevalence of errors was observed for transposed-letter pseudowords compared to the three other conditions. This result can be explained by a stronger lexical capture generated by transposed-letter pseudowords, which are particularly close to words. Moreover, a negative correlation was observed between reading efficiency and error rate across all conditions in the first quantile, indicating that better readers exhibit a higher fast error rate. This characteristic can be explained by the a priori better reading automatism among good readers which makes them more sensitive to the lexical capture.

In conclusion, CAFs analyses computed on EMGs confirmed a stronger lexical capture for transposed-letter pseudowords compared to replaced-letter pseudowords. These findings suggest that the cognitive control mechanisms involved in lexical decision resemble those observed in stimulus-response compatibility tasks, thereby reinforcing the role of control mechanisms in reading activities.

Error-related brain responses are sensitive to error inevitability

Magdalena Senderecka  1, Jakub Szewczyk  1  

1 : Uniwersytet Jagielloński w Krakowie = Jagiellonian University

In everyday life, we usually correctly differentiate between errors that are inevitable and therefore less significant, and those that could be prevented and are thus more important. However, the question arises whether the error-related negativity (ERN), which is an electrophysiological marker of subjective evaluation of error significance, also differentiates between these failures. We hypothesized that the more the control over a motor action is limited by situational constraints, the more reduced the ERN amplitude and the longer the ERN latency should be. In addition, we focused on post-error slowing, which is a post-error behavioral adjustment. We predicted that the more inevitable the error, the smaller the post-error slowing should be. To test these hypotheses, we used the stop signal task, which can be employed to elicit a variety of errors that are avoidable to a greater or lesser extent. The analysis was performed on 55 participants (38 females, mean age 21.5 years) using linear mixed-effects models. Consistent with our hypotheses, we found that the more inevitable the error, the more reduced and delayed the ERN. Moreover, more inevitable errors led to smaller adjustments in subsequent behavior. These results indicate that the elementary basis for distinguishing between avoidable and unavoidable errors may occur in the brain several dozen milliseconds after error commission.

Error-related brain responses are sensitive to error inevitability

Magdalena Senderecka  1, Jakub Szewczyk  1  

1 : Uniwersytet Jagielloński w Krakowie = Jagiellonian University

In everyday life, we usually correctly differentiate between errors that are inevitable and therefore less significant, and those that could be prevented and are thus more important. However, the question arises whether the error-related negativity (ERN), which is an electrophysiological marker of subjective evaluation of error significance, also differentiates between these failures. We hypothesized that the more the control over a motor action is limited by situational constraints, the more reduced the ERN amplitude and the longer the ERN latency should be. In addition, we focused on post-error slowing, which is a post-error behavioral adjustment. We predicted that the more inevitable the error, the smaller the post-error slowing should be. To test these hypotheses, we used the stop signal task, which can be employed to elicit a variety of errors that are avoidable to a greater or lesser extent. The analysis was performed on 55 participants (38 females, mean age 21.5 years) using linear mixed-effects models. Consistent with our hypotheses, we found that the more inevitable the error, the more reduced and delayed the ERN. Moreover, more inevitable errors led to smaller adjustments in subsequent behavior. These results indicate that the elementary basis for distinguishing between avoidable and unavoidable errors may occur in the brain several dozen milliseconds after error commission.

How is deliberation related to volition? 

Guillaume Pech  1, 2, Elisabeth Pacherie  3  , Emilie Caspar  4  , Cleeremans Axel  1  , Uri Maoz  5  

1 : Center for Research in Cognition & Neurosciences [Brussels]

2 : Institute for Interdisciplinary Brain and Behavioral Sciences [Chapman University]

3 : Ecole Normale Supérieure Paris-Saclay

institut jean nicod

4 : Department of Experimental Psychology

5 : Brain Institute Chapman

Neuroscience of volition historically focused on arbitrary decisions but is now challenged to differentiate them from meaningfull ones. Even more importantly as an action is largely considered as more volition when it have been made following a reason. The proposal aims to distinguish between meaningfull and meaningless decisions and between low or high level of deliberation preceding them. Utilizing the Readiness Potential and Temporal Binding —previously linked to volition—as well as self-reported Sense of Volition and the Effort Exerted, the research endeavors to assess the influence of meaningfulness and deliberation on these measures.

We recruited 50 participants, who were instructed to decide between two options to maximize monetary reward. We contrasted the goals from not rewarding and meaningless decisions (i.e., arbitrary condition), to rewarding and meaningful decisions (i.e., deliberate conditions) and we also contrasted the degree of deliberation to decide in an easy deliberate condition and in a hard deliberate condition. We observed that varying the goals of the decision from arbitrary to deliberate decisions impacted several measures related to volition. Participants reported higher Sense of Volition, exhibited a stronger Readiness Potential (RP), have an increase of Temporal Binding (TB) and Effort Exerted (EE) in the deliberate conditions compared to the arbitrary condition. Finally, our results indicate that the degree of deliberation to make the decision did not impact volition. We indeed observed evidence for similar reported volition, RP, TB and EE in the easy deliberate condition and in the hard deliberate condition.

To conclude, our results seem to indicate that the goals of a decision influence the volitional aspect of the action, in term of self-reported feeling of volition (i.e., the SoV), brain activity with the RP, implicit mechanisms with the TB and behavior with the EE. Interestingly, all our measurements indicate similarity between different levels of deliberation. Our results suggest that it is not the complexity of the deliberation prior to the action that makes it more volitional but the goal it deserves. Further, in the present study, we also introduced a new implicit measurement of volition which is the EE measured via the handgrips. This measure was successful to capture what the other measures revealed in term of volition and could thus initiate a novel research line to measure volition. Finally, our study highlights the importance to be cautious on extending previous findings on meaningless decisions to meaningful decisions studying volition.

Insights into reading skills assessment in early literacy: unveiling error patterns in visual word recognition among children via cognitive control-based analyses

Fanny Grisetto  1,  Adrien Lorant  1  , Clémence Roger  1  , Gwendoline Mahé  1  

1 : Laboratoire Sciences Cognitives et Sciences Affectives - UMR 9193

Université de Lille, Centre National de la Recherche Scientifique

Poor reading skills, without any diagnosis of dyslexia, affect about 20% of the population, despite regular schooling, leading to social issues. Our study aimed to explore the origin of error during visual word recognition, a major component of reading skills, in children during the early years of reading acquisition. We investigated error timing during a lexical decision task performed by 62 children (28 in Grade 1 and 34 in Grade 2), assessing their reading and related skills. Reaction times were analysed by Grade, Condition (word vs. pseudoword), and Performance. Also, a dynamic analysis of accuracy was carried out using Conditional Accuracy Function (CAF) curves. Classically used in the cognitive control field, the CAF quantifies response accuracy as a function of reaction times (RTs) and distinguishes fast errors (i.e., in the shortest RTs) from slow errors (i.e., in the slowest RTs). We also calculated the values of the first CAF slope as an index of response capture during the lexical decision task (i.e., lexical capture). 

Results showed that the Correct-Incorrect RTs difference was positive for the pseudoword condition, but negative for the word condition. Also, we observed that the weaker the reading efficiency, the more negative the RT difference in the word condition. This difference was predicted by letter recognition, phonological awareness, and rapid automatized naming of letters. Regarding the CAF analysis, error distribution differed between conditions, revealing slow errors in word trials and fast errors in pseudoword trials. No Grade effect was observed, probably due to the high variability in reading efficiency within both grades. However, when considering reading skills with direct measures, positive correlations were found with the values of the first slopes of the CAF in the pseudoword condition: the better the reading skills, the stronger the lexical capture. This lexical capture was predicted by phonological awareness and oral comprehension scores.

In conclusion, fast pseudoword errors characterize proficient reading, arising from automatic activation of lexical representations. Slow word errors may reveal poor reading skills due to insufficient lexical activation. Borrowed from the cognitive control literature, detailed analyses of errors during a lexical decision task might objectify reading difficulties in children.

Metacognitive instructions finely tune medial pre-frontal cortex performance monitoring activity.

Nathalie Pavailler  1, Franck Vidal  1  , Boris Burle  1  

1 : Centre de Recherche en Psychologie et Neurosciences

Aix-Marseille Université - AMU, CNRS

Cognitive processes and behavioral performance are inherently variable. While this variability is a key feature for biological systems adaptation, it can also have deleterious effects. This variability must hence be internally monitored to detect potential performance degradation. We show that medial pre-frontal cortex (MPFC) activity, known to play a critical role in such monitoring, can be finely tuned to precisely track the relevant variability and ignore the irrelevant one. In a perceptual choice RT task, participants were asked to either evaluate the time they took, or the force they produced, to respond. While both dimensions were highly variable irrespective of the metacognitive instruction, MPFC activity was specifically sensitive to the variations in the evaluated dimension, but not in the non-relevant one. This double dissociation shows that the performance monitoring system is not sensitive to any variation in performance, but on the contrary, that it can be configured to track performance selectively on a given dimension, according to the ongoing cognitive context.

On the task-specificity of adaptive control: A meta analysis

Daxun Zhu  1, Hannah Kaiser  2  , Karin Bausenhart  2  , David Dignath  2  , Carolin Dudschig  2  , Wim Notebaert  1  , Donna Bryce  3  , Senne Braem  1  

1 : Department of Experimental Psychology, Ghent University

2 : Department of Psychology, University of Tübingen

3 : University of Augsburg

Adaptive control refers to flexible adjustments in control settings in response to conflicting situations, often measured using the congruency sequence effect (CSE), which is the observation that the congruency effect in conflict tasks decreases following incongruent compared to congruent trials. There has been a long-standing debate as to whether CSEs reflect a domain-general or domain-specific process – as often tested by studying CSEs across tasks. One model predicts a U-shaped relation where only highly similar or highly dissimilar tasks would show CSEs across tasks, because only those tasks can be represented or activated in parallel. While there has been an abundance of evidence for CSEs across highly similar tasks, only some recent studies seem to have reported CSEs across highly dissimilar tasks, with some failures to replicate. To investigate this issue more systematically, we conducted a preregistered meta-analysis of all relevant cross-task CSE literature in the past (from 1992 to 2023, 146 experiments, 193 effect size). Our first, preliminary results suggest that the CSE can transfer across tasks, and task dissimilarity has a negative linear effect on the size of cross-task CSEs (i.e., not U-shaped). Interestingly, while most measures of task dissimilarity (e.g., stimulus domain, conflict type, relevant stimulus dimension, irrelevant stimulus dimension, sensory mode, response mode) show this negative impact, dissimilar stimulus identity or response identity may actually help increase, rather than reduce cross-task CSEs.

Perceptual uncertainty and response conflict as predictors of error awareness in the flanker task 

Ivan Ivanchei  1, Andrey Chetverikov  2  

1 : Universite Libre de Bruxelles

2 : University of Bergen

Despite the long history of scientific interest in human errors, the pattern of behavioral and physiological changes accompanying errors is far from clear due to the inevitable confusion between errors and error awareness. Disentangling them requires understanding the mechanisms of error awareness. In five experiments, we study the effects of perceptual uncertainty and response conflict on observers' ability to detect their own errors in a flanker task. In Experiment 1, we found that observers' accuracy ratings are sensitive to perceptual uncertainty but not response conflict. Observers are less able to tell the correct answers from errors when target-to-flankers distance is small compared to larger distance. In contrast, incompatible trials compared to neutral ones are similar in error awareness despite significant drop in performance. This result was replicated in Experiment 2 with more difficult stimuli-response mapping. When we varied target eccentricity in Experiment 3, accuracy ratings were affected by target-to-flankers distance and response conflict in interaction with eccentricity. Observers' ability to discriminate between correct answers and errors decreased with increasing eccentricities, and especially so if the flankers were close and incompatible. We suggest that the results can explained by the Bouma's rule well-known in crowding studies and higher similarity between targets and flankers on incompatible trials compared to neutral ones. Further tests in Experiments 4 and 5 with foveally-presented flankers confirmed the results of the first experiments. The findings raise doubts regarding the importance of the response execution stage in error detection in a flanker task.

Predictive Learning as the Basis of the Testing Effect

Haopeng Chen  1, Cristian Buc Calderon  2  , Tom Verguts  1  

1 : Ghent University

2 : Centro Nacional de Inteligencia Artificial (CENIA)

A robust finding in the declarative memory study is the testing effect, meaning that testing enhances retention more than studying. Emergent frameworks propose fundamental (Hebbian and predictive) learning principles as its basis. Predictive learning posits that learning occurs based on the contrast (error) between a prediction and the feedback on that prediction (prediction error). Here, we propose that in testing (but not studying) scenarios, participants predict potential answers, and its contrast with the subsequent feedback yields a prediction error, which facilitates testing-based learning. To investigate this, we developed an associative memory network incorporating Hebbian and/or predictive learning, together with an experimental design where human participants studied or tested English-Swahili word pairs followed by recognition. Two behavioral experiments revealed strong testing effects. Importantly, model fitting suggested that only models incorporating predictive learning can account for the breadth of data associated with the testing effect. Finally, we carried out the same tasks using fMRI to explore the neural basis of the testing effect. We will describe the behavioral, modelling, and neural results, as well as their implications for declarative memory.

Proactive control strategy as a function of reactive inhibition capacities and impulsiveness: challenging the dual mechanism of control 

Marie Mathé  1, 2  , Fanny Grisetto  1  , Clémence Roger  1  

1 : SCALAB UMR CNRS 9193

Univ. Lille, CNRS, UMR 9193 - SCALab - Sciences Cognitives et Sciences Affectives, F 59000 Lille, France

2 : Psychologie : Interactions, Temps, Emotions, Cognition (PSITEC) - ULR 4072

Univ. Lille, ULR 4072 – PSITEC – Psychologie : Interactions, Temps, Émotions, Cognition, F-59000 Lille, France

According to the dual mechanism of control (DMC), reactive and proactive control are involved in adjusting behaviors when maladapted to the environment. The flexible “balance” between the two mechanisms can be seen as a control strategy. This strategy is assessed by the Proactive Behavioral Index (PBI) derived from performances in the AX-CPT task in which the proactive control strategy is appropriate. The PBI indicates the weight of proactive control over reactive control (a higher PBI indicating a more proactive strategy). Interestingly, inter-individual factors (such as age or cognitive capacities) and contextual factors (such as stressful or motivational situations) influence the preferential control strategy. Specifically, individuals with poor reactive inhibition capacities, thus impeding the efficiency of reactive control, are hypothesized to adopt a more proactive control strategy. We tested this hypothesis while exploring the influence of impulsivity on this reactive-proactive counterbalance.

In the current study, 38 adults performed four blocks of an AX-CPT task and four blocks of a Stop Signal task. The evolution of the PBI across the four AX-CPT blocks was modelized through a regression analysis for each participant, to extract the intercept (i.e., how proactive the participant was initially) and the slope coefficient (i.e., how fast the participant did adapt their control strategy to task demands). The Stop Signal Reaction Time (SSRT) was used to assess reactive inhibition capacities. Additionally, participants filled out the UPPS questionnaire to assess their impulsiveness.

In line with the DMC hypothesis, the analyses revealed that the intercept was positively correlated with the SSRT, suggesting that poorer reactive inhibition capacities were associated with a general tendency towards proactive control strategy. When taking into account impulsiveness in this analysis, the SSRTs tended to be correlated to higher intercept only for high impulsive individuals. Secondarily and expectedly, the analyses on the PBI revealed that the intercept was negatively correlated with the slope coefficient, suggesting that individuals with an initially more reactive control strategy adapted to task demands because their strategy was not already well-aligned with the task's demands at the beginning.

Our findings confirm the counterbalance of poor reactive inhibition capacities by a more proactive control strategy. Surprisingly, it seems to be observed only in the more impulsive individuals of our sample. The implications of this finding for pathologies with an impulsive behavioral characteristic (e.g., binge eating) will be discussed.

Social error monitoring in professional basketball games

Ayala Denul  1,  Wim Notebaert  2, Gilles Pourtois  3 

1 : Department of Experimental Psychology, Ghent University

2 : Department of Experimental Psychology, Ghent University

3 : Department of Experimental, Clinical and Health Psychology, Ghent University

Error monitoring, or detecting and adapting to errors, is a crucial aspect of behavioral regulation which has been studied extensively in cognitive psychology and neuroscience. Lab studies often observe increased caution and slowed down responses following own and others' errors. Moreover, social error monitoring research suggests that cooperation and shared goals elicit stronger adaptations to others' errors than competition. However, these insights primarily stem from controlled reaction time tasks, potentially limiting generalization to real-world social contexts. To address this, we studied social post-error adaptations in competitive basketball by analyzing data from four NBA seasons. Our meta-analyses revealed increased caution in terms of shot distance and angle, following missed compared to made shots with the most pronounced effect after own, next teammates' and last opponents' errors. While controlling for rebounds reduced the effects following own and teammates' errors, the general pattern persisted. Additionally, we found a positive correlation between post-error caution and post-error accuracy, underscoring the adaptive nature of this cautious response. Importantly, our results also revealed a positive correlation between adaptations following own and teammates' errors, contrasting with a negative correlation with adaptations following opponents' errors. This suggests that the effects following own and teammates' errors may partially be driven by similar processes, while different mechanisms may be involved following opponents' errors. These findings not only extend cognitive theories from controlled lab settings to a real-word sport setting, but also provide a crucial understanding of athletes' social error responses thereby paving the way for research aimed at optimizing these responses.

The dual influence of interpersonal bonds on the cognitive processes of disobedience

Nicolas Coucke  1, Salvatore Lo Bue  2  , Daniele Marinazzo  1  , Emilie Caspar  1  

1 : Universiteit Gent

2 : Royal Military Academy (RMA)

Most humans have a strong tendency to obey the orders of superiors, even when these orders may cause harm to others. Previous studies pointed out that prosocial disobedience to such orders might be supported by empathy towards a potential victim. Meanwhile, a close relationship with a superior may increase the tendency to obey orders. We hypothesize that both one's relationship with a potential victim and one's relationship with a commander might modulate the ease with which individuals obey or disobey immoral orders. To explore this, we designed an experimental paradigm where 3 participants were each assigned a fixed role of either victim, agent, or commander. Initially, the agent engaged in a joint drawing task with both the victim and the commander to establish interpersonal relationships. In a subsequent phase, the agent was instructed to either obey or disobey the commander's orders to administer moderately painful shocks to the victim. Simultaneous EEG recordings of the 3 participants were made during the entire experiment. Preliminary results indicate that cognitive conflict, as indicated by mid-frontal theta activity, during antisocial obedience is positively correlated with the subjective strength of the agent-victim relationship, while cognitive conflict during prosocial disobedience is positively correlated to the agent-commander relationship. These findings shed light on how complex social situations influence cognitive processes related to obedience and disobedience

The Two Models Of Cognitive Effort: Experimental Evidence Favour Active Arousal Over Negative Valence

Ali Adeli Koudehi  1, 2, 3, 4

1 : ali adeli Koudehi

LNC - Aix-Marseille Université

2 : Marie-Helene Grosbras

Aix-Marseille Université - AMU

3 : Stephane Luchini

Aix Marseille School of Economics

4 : Miriam Teschl

Ecole des Hautes Etudes en Sciences Sociales (EHESS)

This paper provides an overview of the concept of cognitive effort within the realm of neuroeconomics and discusses methods for its quantification. It surveys existing literature on effort and cognitive effort, delineating two predominant perspectives: active arousal and negative valence. These perspectives offer distinct predictions and suggest varied approaches to measuring cognitive effort and its relationship with performance. To adjudicate between these perspectives, we present a novel experimental design incorporating pupillometry and electroencephalography (EEG) data acquisition. Findings lend support to the arousal perspective of cognitive effort. 

Unconscious processing of monetary rewards – a failed replication

Pietro Amerio  1, Guillaume Pech  1, 2, Axel Cleeremans  1  

1 : Center for Research in Cognition & Neurosciences [Brussels]

2 : Institute for Interdisciplinary Brain and Behavioral Sciences [Chapman University]

Expected rewards are known to influence how much effort is put in performing a task. This is a key mechanism of motivation. Several studies reported that unconscious rewards can drive behavior, such that motivation for action is not necessarily a conscious evaluation process. In this work, we take into consideration a seminal study on the topic by Pessiglione and colleagues (2007). They were the first to show that subliminal monetary rewards influence the amount of effort that participants are willing to invest. In the original study, masked pictures of either one pound or one penny coins were shown for variable durations. Participants were instructed that they would be rewarded a proportion of the presented coin (1£ or 1p) matching the amount of force they exerted on a handgrip. For all stimulus durations (17, 50 and 100 ms), the exerted force was modulated by the reward at stake. Crucially, in a separate task, participants reported on the identity and on their subjective visibility of the same kind of stimuli. This enabled the authors to infer that stimuli of 50ms or shorter were processed without awareness and to conclude that force exertion could be motivated even by subliminal rewards. Similar effects were shown for galvanic skin responses (GSR; a known marker of autonomic activation) and BOLD responses in the ventral pallidum (reportedly involved in reward processing). The Pessiglione et al. (2007) study holds great relevance for theories of consciousness and motivation, but its results are available to criticism. In particular, testing for awareness outside the main (incentive force) task does not guarantee reliable awareness reports. Additionally, the analyses did not account for individual differences, which could lead to biased results, especially given the limited sample size and number of trials per condition. We set out to replicate the results from Pessiglione et al. (2007), first by totally adhering to their experimental design and analyses, and, in a successive step, by ameliorating the methodology using state-of-the-art protocols. At the conference, we will present results from the first study, which faithfully reproduced the original experiment, replacing pounds with euros to ensure that our participants were familiar with their appearance and value. The only other difference is that we recorded brain activity using EEG (instead of fMRI) and tested the effect of conscious vs. unconscious monetary stakes on the N2 and P3 components (reported markers of motivation processing). We could not find behavioral evidence that masked euros and cents could be perceived outside awareness, nor that more grip strength was applied to euros (vs. cents) when subjects were unaware of them. Likewise, all our physiological measures (GSR, N2, P3) were affected by the monetary stakes only in conditions in which stimuli were consciously processed. Crucially, the same results were obtained regardless of whether we adopted a subjective or an objective criterion for establishing awareness of the stimuli. These results call for a deeper inquiry of claims of unconscious reward processing and motivation.