100 research outputs found
Data from Filevich, Horn, and Kühn (2017) Psychological Research: Within-person adaptivity in frugal judgments from memory
No full textData from Filevich, Horn, & Kühn (2017). Psychological Research.
Each line includes the data from one participant in the study. Columns and category numbers refer to the MPT model categories as described in the main article, for the population and distance condition, respectively.
If you use these data in your research, please refer to the article as: Filevich, E., Horn, S. S., & Kühn, S. (2017). Within-person adaptivity in frugal judgments from memory. Psychological Research
Prior information differentially affects discrimination decisions and subjective confidence reports
No full textContributions of tactile information to the sense of agency and its metacognitive representations
No full textMetacognition Across Different Modalities in Health and Disease
No full textMetakognition ermöglicht Einblicke in eigene kognitive Prozesse und beeinflusst menschliches Verhalten in mehreren Bereichen. Die Debatte, ob Metakognition sich domänenallgemein oder domänenspezifisch auswirkt, bleibt unentschieden, wobei die meisten Studien auf sensorische Informationsverarbeitung und Gedächtniseffekte fokussieren. Diese Dissertation unterstreicht die Notwendigkeit, weitere Domänen zu untersuchen, um diese Frage zu klären, und tut dies, indem sie Metakognition im motorischen Bereich erforscht.
Diese Arbeit umfasst vier empirische Studien. Studie 1 untersuchte die zugrundeliegenden Strukturen in metakognitiven Domänen und fand keine Beweise, die eine Unterscheidung zwischen intern oder extern generierter Information unterstützen. Studie 2 konzentrierte sich auf motorische Metakognition und verglich die Introspektion direkter und indirekter Bewegungsparameter. Obwohl beide Bewegungsparameter gleichermaßen introspektierbar sind, weisen fehlende Korrelationen zwischen verschiedenen Bewegungsparametern innerhalb einer Domäne auf die Rolle von nicht-domänenspezifischen Merkmalen in der Metakognition hin. In einem datengesteuerten Ansatz zeigte Studie 3, wie die Introspektion verschiedener Aspekte der Bewegung mit motorischer Kontrolle und Metakognition verknüpft ist. In Studie 4 wurde getestet, ob metakognitive Defizite die Vorgefühle bei Tourette-Störung untermauern. Es fanden sich keine Belege für Beeinträchtigungen in taktiler oder visueller Metakognition, was die Rolle der Metakognition in der Pathophysiologie der Störung in Frage stellt.
Insgesamt betont die Arbeit die Notwendigkeit, domänenbezogene und aufgabenspezifische Merkmale zu entflechten und Faktoren zu identifizieren, die Allgemeinheit und Spezifität beeinflussen. Sie trägt zum Verständnis der motorischen Metakognition bei, indem sie experimentelle und analytische Ansätze vorantreibt und gleichzeitig Einblicke in die Architektur höherer kognitiver Funktionen und das Bewusstsein bietet.Metacognition, the ability to introspect into one's cognitive processes, can guide human behaviour across various domains. The debate whether metacognition operates in a more domain-general or a domain-specific fashion remains inconclusive, with most studies focused on sensory and memory domains. This thesis argues that to resolve this question, it is necessary to broaden the scope of domains, and it does so by exploring metacognition in the motor domain.
This thesis comprises four empirical studies. Study 1 examined the underlying structures in metacognitive domains, finding no evidence to support a distinction between internally or externally generated information. Study 2 focused on motor metacognition, comparing the monitoring of direct and indirect movement parameters. While both movement parameters can be equally well monitored, the absence of correlations within the same domain but across different monitored parameters highlights the role of non-domain features on metacognition. In a data-driven approach, Study 3 revealed how monitoring different aspects of movement is linked with motor control and metacognitive judgments. Study 4 tested whether metacognitive deficits underpin premonitory urges in Tourette's disorder. No evidence of metacognitive impairment was found in the tactile or visual domains, challenging metacognition's role in the disorder's pathophysiology.
Overall, the thesis emphasises the need to disentangle domain-related and task-specific features, advocating for identifying factors that influence generality and specificity. It contributes to the understanding of motor metacognition by advancing experimental and analytical approaches, while providing insights into the architecture of higher cognitive functions and conscious processing
Metacognition of motor imagery
No full textMotor metacognition refers to the knowledge about one’s actions and movements. Investigating motor metacognition is complex because, in principle, many aspects of a movement might be available to metacognitive monitoring. These aspects include movement intentions, motor preparation, motor execution, the sensory consequences that result from movement, as well as the sense of agency. Here we aim to tease apart some of these components and study their potential contribution to the metacognitive monitoring of movements. In particular, we will focus on motor imagery, the process of imagining movements without executing them (Fuchs et al., 2020). Motor imagery separates afferent from efferent processes informing motor metacognition, as metacognitive judgments about motor imagery (e.g., vividness) can only rely on efferent information, as no afferent sensory information is available since movements are not executed.
In order to study metacognition of motor imagery, we will rely on the vast research that has been performed on the metacognition of visual imagery. In a seminal paper, Pearson and colleagues (2011) investigated whether metacognitive vividness ratings of visual imagery predict perceptual consequences of that imagery in a binocular rivalry task. Results showed that on trials where participants reported highly vivid imagery, participants had a stronger tendency towards perceiving the imagined visual pattern, as compared to less vivid trials. Additionally, participants with a greater general visual imagery ability showed a stronger perceptual biasing effect than participants with less visual imagery ability (Pearson et al., 2011). Here, we aimed at expanding these findings to the motor domain. This will reveal whether young healthy adults can metacognitively introspect their own motor imagery, and provide a means to further investigate interindividual differences and the neural mechanisms that this potential ability relies on.
This preregistration is a modified version of an existing preregistration (10.17605/OSF.IO/KC2M7). Preliminary analysis of the data collected in line with that preregistration did not reveal the expected interaction between subjective vividness and the strength of the congruency effect. Potentially, the two possible movements used are too similar to each other, so that imagination of one movement facilitates the execution of the other movement. Therefore, we will modify the movements from a power and a precision grip to a crimp grip and a sideway pull, based on the findings of Bläsing and colleagues (2014). Based on pilots and experiences with the previous study, changes are introduced in the experimental motor imagery task and the movement apparatus
Behavioural, modeling, and electrophysiological evidence for supramodality in human metacognition
No full textThree very comparable tasks, in three different perceptual domains: vision, audition and touch. We ask: is metacognitive ability stable across domains? Does multisensoy information improve metacognition, beyond performance? If so, how
Behavioural, modeling, and electrophysiological evidence for supramodality in human metacognition
No full textThree very comparable tasks, in three different perceptual domains: vision, audition and touch. We ask: is metacognitive ability stable across domains? Does multisensoy information improve metacognition, beyond performance? If so, how
The Role of Priors in Confidence
No full textAccording to an extensive literature on predictive processing and Bayesian inference, our perception depends on the integration of prior expectations or constraints (priors) and the incoming information (likelihood), to form the posterior. Building off of this Bayesian framework, standard models have considered confidence to be based on the perceived posterior probability that a decision is correct, given the internal evidence and the decision (Sanders et al., 2016; Pouget et al., 2016; Fleming & Daw, 2017). This assumes that confidence is based on the optimal integration of priors and likelihoods. However, this may not be the case, and in order to make quantitative predictions about confidence across different situations and rigorously test confidence models, it is critical to understand how priors are integrated.
While some recent work has shown a general influence of priors on confidence (Sherman et al., 2015), an examination of whether confidence weights priors and likelihoods equally and optimally in this way is lacking. Additionally, recent work has revealed a variety of suboptimalities in confidence computations, including some suggesting general overweighting of sensory precision (Navajas et al., 2017; Spence et al., 2015; Gardelle & Mamassian, 2015), and some revealing suboptimalities in the incorporation of priors and payoffs into confidence (Locke et al., 2020). This points further to the possibility that our confidence and metacognitive monitoring of information from priors versus incoming sensory information may differ. Here, we will examine this directly by varying the precision of priors and likelihoods and examining whether perceptual decisions and confidence differ depending on whether priors or likelihoods are more informative.
We will use a dual-decision task involving two moving dot stimuli - one prior stimulus, forming the prior, and one target stimulus, forming the likelihood. Participants will make two left/right decisions on each trial, to report whether they think that the main direction of dot movement for each of the prior and target stimuli is rightwards or leftwards. They will then rate their confidence in their response about the target. The correct direction of the target stimulus will be determined by the response accuracy of the prior decision (similar to the task used by Lisi et al, 2020), such that if participants are correct about the prior decision, the target stimulus will move to the right, and if they are incorrect about the prior decision, the target stimulus will move to the left. Participants will be informed of this task structure by way of a gamified task, in which the moving dots will represent a flock of sheep, and they will first need to send their sheepdog to herd them. If their sheepdog is in the correct place (i.e. their first decision was correct) the sheep will be herded toward the barn and will go to the right on the next stimulus. If not (i.e. their first decision was incorrect), the sheep will run away to the fields to the left. This way, participants’ prior for “right” during the target stimulus should be equal to their confidence that they sent their sheepdog to the correct place, or their decision confidence about the prior stimulus. After viewing the target stimulus, they will send their farmer to get their sheep, either from the barn or fields (right or left), and will rate their confidence that their farmer got the sheep. This confidence would optimally be informed by both the prior and target stimuli. By having the prior and target come from the same stimulus type, we can vary the precision of the prior and likelihood on the same axis, to create two conditions with equal posterior precisions:
(i) Condition 1: Precise-Prior - the prior is the more precise and hence informative cue, despite the posterior precision being equal to Condition 2
(ii) Condition 2: Precise-Likelihood - the likelihood is the more precise and hence informative cue, despite the posterior precision being equal to Condition 1
We will then compare decision parameters and confidence between the two conditions, to assess whether priors and targets are weighted equally
Tactile Meta-Agency: How does tactile information influence our metacognition of agency?
No full textWhen we intentionally move our body, our brain is assumed to compare the sensory predictions to the sensory consequences accompanying our actions. If these two match, then we feel that we are the agent of our actions. However, if there is a discrepancy between action predictions and the sensory consequences we attribute the actions to an external force (e.g. we feel that our hand was moved by someone else). Empirical studies often investigate this sense of agency (SoA) by manipulating primarily visual representations of either the movement or the consequences of that movement on the environment.
While most experimental manipulations focus on visual consequences of movement, tactile information can, and does, also guide our movements by encoding properties of the external world. To investigate the contributions of tactile information to representations of our own movements, in the present experiment we will use a novel experimental design to investigate whether visuo-tactile information related to the outcome of a movement affects our SoA similarly to visuo-proprioceptive information alone. We will test participants in two tasks, one where we will measure subjective ratings of agency when different aspects of the movement displayed (spatial, temporal, tactile) will be manipulated, and one where we will measure their metacognition of agency using the very same manipulations.
In the first task (Agency-task), participants will move their right hand toward one of two possible targets. Each target is a circular ridged plate rotated either clockwise or counterclockwise from the vertical axis. We will track participants’ movements with sensors placed on their hand and will use this tracking information to present a virtual hand in real-time on the screen. We will include trials where the visual feedback will be manipulated in one of three possible ways. In three conditions, either the timing of the movement, the location of the hand in virtual space, or the direction of the ridges on the physical plate will be incongruent with the real movement or tactile sensation. For each condition, we will use five levels of magnitude manipulation. As it is often done in agency tasks, participants will rate how much agency they felt they had over the virtual hand, at the end of each movement.
In the second task (MetAgency-task), rather than measuring subjective ratings we will quantify participants’ metacognition of agency, to compare the different conditions devoid of the confounds and report biases that are intrinsic to the subjective ratings used in the former task. We will integrate the agency task into a two-interval forced choice task. In this task, participants will move their right hand towards the target two consecutive times. Participants first will discriminate over which of the two intervals they felt more in control and then they will rate their confidence in their own decision. To compare participants' metacognition of agency we will match participants’ first-order performance using online staircase procedures. For each task, we will test each condition in separate blocks of trials which will be pseudorandomised and counterbalanced.. We will complete data collection in two separate sessions for each subject
Tactile Meta-Agency: How does tactile information influence our metacognition of agency?
No full textWhen we intentionally move our body, our brain is assumed to compare the sensory predictions to the sensory consequences accompanying our actions. If these two match, then we feel that we are the agent of our actions. However, if there is a discrepancy between action predictions and the sensory consequences we attribute the actions to an external force (e.g. we feel that our hand was moved by someone else). Empirical studies often investigate this sense of agency (SoA) by manipulating primarily visual representations of either the movement or the consequences of that movement on the environment.
While most experimental manipulations focus on visual consequences of movement, tactile information can, and does, also guide our movements by encoding properties of the external world. To investigate the contributions of tactile information to representations of our own movements, in the present experiment we will use a novel experimental design to investigate whether visuo-tactile information related to the outcome of a movement affects our SoA similarly to visuo-proprioceptive information alone. We will test participants in two tasks, one where we will measure subjective ratings of agency when different aspects of the movement displayed (spatial, temporal, tactile) will be manipulated, and one where we will measure their metacognition of agency using the very same manipulations.
In the first task (Agency-task), participants will move their right hand toward one of two possible targets. Each target is a circular ridged plate rotated either clockwise or counterclockwise from the vertical axis. We will track participants’ movements with sensors placed on their hand and will use this tracking information to present a virtual hand in real-time on the screen. We will include trials where the visual feedback will be manipulated in one of three possible ways. In three conditions, either the timing of the movement, the location of the hand in virtual space, or the direction of the ridges on the physical plate will be incongruent with the real movement or tactile sensation. For each condition, we will use five levels of magnitude manipulation. As it is often done in agency tasks, participants will rate how much agency they felt they had over the virtual hand, at the end of each movement.
In the second task (MetAgency-task), rather than measuring subjective ratings we will quantify participants’ metacognition of agency, to compare the different conditions devoid of the confounds and report biases that are intrinsic to the subjective ratings used in the former task. We will integrate the agency task into a two-interval forced choice task. In this task, participants will move their right hand towards the target two consecutive times. Participants first will discriminate over which of the two intervals they felt more in control and then they will rate their confidence in their own decision. To compare participants' metacognition of agency we will match participants’ first-order performance using online staircase procedures. For each task, we will test each condition in separate blocks of trials which will be pseudorandomised and counterbalanced.. We will complete data collection in two separate sessions for each subject
- …
