Fechner’s law in metacognition: A quantitative model of visual working memory confidence.

Berg, Ronald van den; Yoo, Aspen H.; Ji, Wei

doi:10.1037/rev0000060

Cited by 92 publications

(102 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At the same time, many other experiments imply the presence of dissociable neural circuits for perceptual decision making and confidence. Behaviorally, confidence judgments can be dissociated from the accuracy of the perceptual decision in both humans (Lau and Passingham, 2006;Rahnev et al, 2011bRahnev et al, , 2015Zylberberg et al, 2012Zylberberg et al, , 2016Vlassova et al, 2014;Spence et al, 2015;Koizumi et al, 2015;Song et al, 2015;Samaha et al, 2016Samaha et al, , 2017Boldt et al, 2017;Peters et al, 2017;Desender et al, 2018) and monkeys (Ferrigno et al, 2017), whereas other studies suggest that confidence judgments but not perceptual decisions are subject to late metacognitive noise (Mueller and Weidemann, 2008;Jang et al, 2012;De Martino et al, 2013;Maniscalco and Lau, 2016;Rahnev et al, 2016;van den Berg et al, 2017;Shekhar and Rahnev, 2018;Bang et al, 2019). Neurally, studies employing transcranial magnetic stimulation (TMS) delivered to the prefrontal cortex have been able to alter subjects' confidence ratings, while leaving their perceptual decisions unaffected (Rounis et al, 2010;Fleming et al, 2015;Rahnev et al, 2016;Shekhar and Rahnev, 2018).…”

Section: Introductionmentioning

confidence: 99%

Overlapping and unique neural circuits are activated during perceptual decision making and confidence

Yeon

Shekhar

Rahnev

2018

Preprint

View full text Add to dashboard Cite

Word count: 178)Perceptual decisions are naturally accompanied by a sense of confidence in the accuracy of the decision. However, it remains unclear whether perceptual decision making and confidence are supported by the same or different neural circuits. To address this question, we conducted two functional MRI (fMRI) experiments in which we dissociated the periods related to perceptual decision making and confidence by either decorrelating their respective regressors or asking for confidence ratings only in the second half of the experiment. We found that perceptual decision making and confidence were supported by large and mostly overlapping brain circuits including frontal, parietal, posterior, and cingulate regions with the results being remarkably consistent across the two experiments. Further, the confidence period recruited a number of unique regions, whereas there was no evidence for the decision period recruiting unique regions not involved in the confidence period. These results suggest that the neural circuits supporting perceptual decision making and confidence have a very high degree of overlap, which suggests largely shared computational mechanisms with unique components for the confidence but not for the perceptual judgments.

show abstract

Section: Introductionmentioning

confidence: 99%

Overlapping and unique neural circuits are activated during perceptual decision making and confidence

Yeon

Shekhar

Rahnev

2018

Preprint

View full text Add to dashboard Cite

show abstract

“…Beyond this, people may have knowledge of their uncertainty on a trial-to-trial level: when people make explicit reports of confidence in memory decisions, the amount of response error correlates with the reported confidence on each trial 10;12 . This could be explained by memory confidence being a function of internal fluctuations in underlying memory quality 13 . However, confidence ratings are not necessarily a reflection of memory uncertainty 14;15 and may be produced through a different mechanism from those used to make decisions under uncertainty 16;17 .Thus, to observe memory uncertainty, experimenters may not want to rely on explicit reports of uncertainty alone, but also use paradigms in which people are incentivized to make decisions which implicitly incorporate memory uncertainty by combining it with other information.…”

mentioning

confidence: 99%

Humans incorporate trial-to-trial working memory uncertainty into rewarded decisions

Honig

Fougnie

2018

Preprint

Self Cite

View full text Add to dashboard Cite

In daily life, working memory plays an important role in action planning and decision-making. However, both the informational content of memory and how that information is used in decisions is poorly understood. To investigate this, we conducted a memory experiment where people not only reported an estimate of a remembered stimulus, but also made a rewarded decision designed to reflect memory uncertainty. Reported memory uncertainty is correlated with estimation error, showing people incorporate their trial-to-trial memory quality into rewarded decisions. Moreover, memory uncertainty can be combined with other sources of information; after we induced prior beliefs about stimuli probabilities, we found that estimates shifted towards more frequent colors, with the shift increasing with reported uncertainty. The data is best predicted by models where people incorporate their trial-to-trial memory uncertainty with potential rewards and prior beliefs, highlighting the importance of studying working memory as a process integrated with decision-making.Working memory (WM), the storage and manipulation of information on a short timescale, is essential for many cognitive processes. For instance, individual differences in WM predict intelligence and academic success 1-3 . To understand WM, research has largely focused on examining the capacity and limitations of WM (reviewed in Luck and Vogel , 2013 4 , Ma et al. , 2014 5 ), epitomized by memory paradigms such as delayed estimation 6-8 , in which participants report a guess of a stimulus feature after a delay. In real life, however, WM information is not only used to make estimates of stimuli features, but to make decisions and take actions. For example, when deciding when to cross the street, a person must remember from a glance the position and velocity of cars. Since a mistake in this decision is costly, and memories are noisy, WM information ought to be combined with information about potential rewards (getting to your destination sooner) and costs (getting hit by a car) of the decision. In doing so, it is useful to know how reliable, or conversely how uncertain, one's memory is. If someone is uncertain about the speed of an oncoming car, they might play it safe and wait a little longer in order to avoid a high-cost collision.Yet uncertainty in working memory is rarely studied. There is evidence that people know something about the quality of their WM representations 9-12 . For example, people can report which items from a set of stimuli they remember better 9 . Beyond this, people may have knowledge of their uncertainty on a trial-to-trial level: when people make explicit reports of confidence in memory decisions, the amount of response error correlates with the reported confidence on each trial 10;12 . This could be explained by memory confidence being a function of internal fluctuations in underlying memory quality 13 . However, confidence ratings are not necessarily a reflection of memory uncertainty 14;15 and may be produced through a different mechanism from those...

show abstract

“…However, this experiment, along with much of the VWM literature, overlooks other information available in VWM: memory uncertainty. Indeed, people can successfully report on the quality of their memory-based decisions (Rademaker, Tredway, & Tong, 2012), suggesting a representation and use of uncertainty over the memorized stimulus (Fougnie, Suchow, & Alvarez, 2011;Honig, Ma, & Fougnie, 2018;van den Berg, Yoo, & Ma, 2017). We conducted a second experiment to investigate how, if at all, priority affects working memory uncertainty.…”

mentioning

confidence: 99%

Strategic allocation of working memory resource

Yoo

Klyszejko

Curtis

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

Visual working memory (VWM), the brief retention of past visual information, supports a range of cognitive functions (Fukuda, Vogel, Mayr, & Awh, 2010; Johnson et al., 2013).The resource that supports VWM is limited, raising the question of how the brain allocates this limited resource to different objects. This question is even more interesting in ecological settings, in which objects are not equally important. In a psychophysical experiment, participants remembered the location of four targets with different probabilities of being tested after a delay. We then measured their memory accuracy of one of the targets. We found that participants allocated more resource to memoranda with higher priority, but underallocated resource to high-and overallocated to low-priority targets relative to the true probe probabilities. These results are well explained by a computational model in which resource is allocated to minimize expected estimation error. We replicated this finding in a second experiment in which participants bet on their memory fidelity after making the location estimate. The results of this experiment show that people 1) use information about memory quality and 2) minimize error even with an incentivized, alternative resource allocation strategy. Humans may mitigate the behavioral effects of a limited VWM through knowledge of memory fidelity and strategic resource allocation.

show abstract

Fechner’s law in metacognition: A quantitative model of visual working memory confidence.

Cited by 92 publications

References 68 publications

Overlapping and unique neural circuits are activated during perceptual decision making and confidence

Overlapping and unique neural circuits are activated during perceptual decision making and confidence

Humans incorporate trial-to-trial working memory uncertainty into rewarded decisions

Strategic allocation of working memory resource

Contact Info

Product

Resources

About