Higher Meta-cognitive Ability Predicts Less Reliance on Over Confident Habitual Learning System

Ershadmanesh, Sara; Miandari, M; Vahabie, Abdol-Hossein; Mn, Ahmadabadi

doi:10.1101/650556

Cited by 3 publications

(4 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…People with lower average confidence had a weaker model-free contribution to their decisions. This is consistent with findings in Ershadmanesh et al (2019), where authors showed that there is a negative correlation between the model-free component of decision and memory meta-metacognition. Since the two-step decision-making task is designed in a way that you could not have a better than random performance (Akam et al, 2015;Kool et al, 2016), it can be argued that people with lower confidence understood this fact better and were more meta-cognitive.…”

Section: Discussionsupporting

confidence: 92%

“…The above proposal is in line with Frith and Frith (2012)'s discussion about exertion of control over automatic behavior via meta-cognitive processes. Ershadmanesh et al (2019)'s result confirm our proposal, too. They show a positive relationship between the proportion of model-based relative to model-free behavior and self-monitoring ability.…”

Section: Introductionsupporting

confidence: 89%

See 1 more Smart Citation

The effects of directed and free self-monitoring on goal-directed and habitual decision-making

Miandari Hossein,

Ershadmanesh,

Vahabie

et al. 2023

Preprint

View full text Add to dashboard Cite

Many studies on humans and animals have provided evidence for the contribution of goal-directed and habitual valuation systems in learning and decision-making. However, how the arbitration between these two systems is affected by other cognitive processes is not well known. Here, we study the effects of directed and free self-monitoring of one’s decisions on this arbitration. In our experiments, in a within-subject design, the subjects participated in a control and a two modified versions of the Two-step decision-making task, where we could measure each system’s contribution to decisions. We had two modified tasks. In both, every few trials subjects had to think about what they have experienced in the past trials in one of the two days. In one task, they had to designate which action was better and then report their confidence about this decision (directed self-monitoring task). In the other modification of task, they had to explain what had happened in the past few trials by talking (free self-monitoring task). We hypnotized that in both modified tasks, the behavior of the participants would shift toward goal-directed behavior because they need to think more about the structure of the task. Our experimental results showed that subjects indeed became more goal-directed in the directed self-monitoring task, but in the free self-monitoring task, they became more habitual. We would discuss the underlying reasons for these shifts in the behavior.

show abstract

Section: Discussionsupporting

confidence: 92%

Section: Introductionsupporting

confidence: 89%

The effects of directed and free self-monitoring on goal-directed and habitual decision-making

Miandari Hossein,

Ershadmanesh,

Vahabie

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…These findings clarify prior work identifying an increase in metacognitive bias in aversive contexts [29], and provide support for our assertion that approach and avoidance motivations underlie differences previously attributed to context. Because our design improved accuracy-demand trade-offs through deterministic state transitions, our results are also consistent with recent work showing higher decision confidence is associated with model-free learning when model-based and model-free systems have chance level performances [30]. In this study, reduced metacognitive bias was associated with more model-based control and faster learning rates.…”

Section: Plos Computational Biologysupporting

confidence: 91%

Model-based prioritization for acquiring protection

2022

View full text Add to dashboard Cite

Protection often involves the capacity to prospectively plan the actions needed to mitigate harm. The computational architecture of decisions involving protection remains unclear, as well as whether these decisions differ from other beneficial prospective actions such as reward acquisition. Here we compare protection acquisition to reward acquisition and punishment avoidance to examine overlapping and distinct features across the three action types. Protection acquisition is positively valenced similar to reward. For both protection and reward, the more the actor gains, the more benefit. However, reward and protection occur in different contexts, with protection existing in aversive contexts. Punishment avoidance also occurs in aversive contexts, but differs from protection because punishment is negatively valenced and motivates avoidance. Across three independent studies (Total N = 600) we applied computational modeling to examine model-based reinforcement learning for protection, reward, and punishment in humans. Decisions motivated by acquiring protection evoked a higher degree of model-based control than acquiring reward or avoiding punishment, with no significant differences in learning rate. The context-valence asymmetry characteristic of protection increased deployment of flexible decision strategies, suggesting model-based control depends on the context in which outcomes are encountered as well as the valence of the outcome.

show abstract

“…Because we chose to improve accuracy-demand trade-offs through deterministic state transitions, our results are also consistent with recent work showing higher decision confidence is associated with model-free learning when model-based and model-free systems have chance level performances. 27 In this study, better metacognitive accuracy was associated with more model-based control and faster learning rates. Thus, model-based actions can be interpreted to reflect explicit forecasts.…”

Section: Discussionmentioning

confidence: 51%

Model-based prioritization for acquiring protection

Tashjian¹,

Wise²,

Mobbs³

2021

Preprint

View full text Add to dashboard Cite

Protection, or the mitigation of harm, often involves the capacity to prospectively plan the actions needed to combat a threat. The computational architecture of decisions involving protection remains unclear, as well as whether these decisions differ from other positive prospective actions. Here we examine effects of valence and context by comparing protection to reward, which occurs in a different context but is also positively valenced, and punishment, which also occurs in an aversive context but differs in valence. We applied computational modeling across three independent studies (Total N=600) using five iterations of a ‘two-step’ behavioral task to examine model-based reinforcement learning for protection, reward, and punishment in humans. Decisions motivated by acquiring safety via protection evoked a higher degree of model-based control than acquiring reward and avoiding punishment, with no significant differences in learning rate. The context-valence asymmetry characteristic of protection increased deployment of flexible decision strategies, suggesting model-based control depends on the context in which outcomes are encountered as well as the valence of the outcome.

show abstract

Higher Meta-cognitive Ability Predicts Less Reliance on Over Confident Habitual Learning System

Cited by 3 publications

References 29 publications

The effects of directed and free self-monitoring on goal-directed and habitual decision-making

The effects of directed and free self-monitoring on goal-directed and habitual decision-making

Model-based prioritization for acquiring protection

Model-based prioritization for acquiring protection

Contact Info

Product

Resources

About