Implicit learning mediates base rate acquisition in perceptual categorization

Bohil, Corey J.; Wismer, Andrew

doi:10.3758/s13423-014-0694-2

Cited by 13 publications

(24 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, there are many everyday tasks in which the probability of a set of alternatives needs to be assessed based on one’s past experience with the outcomes of the task. The importance of experience has been demonstrated in studies examining differences between experience-based and description-based decisions [10, 11] and in perceptual-categorization tasks with unequal probability, in which response feedback leads to performance that is closer to optimal than observational feedback [12, 13]. While these studies demonstrate the importance of experience on decision-making behavior, they do not describe how experience influences expectation formation.…”

Section: Introductionmentioning

confidence: 99%

Human online adaptation to changes in prior probability

Norton

Acerbi

et al. 2019

PLoS Comput Biol

View full text Add to dashboard Cite

Optimal sensory decision-making requires the combination of uncertain sensory signals with prior expectations. The effect of prior probability is often described as a shift in the decision criterion. Can observers track sudden changes in probability? To answer this question, we used a change-point detection paradigm that is frequently used to examine behavior in changing environments. In a pair of orientation-categorization tasks, we investigated the effects of changing probabilities on decision-making. In both tasks, category probability was updated using a sample-and-hold procedure: probability was held constant for a period of time before jumping to another probability state that was randomly selected from a predetermined set of probability states. We developed an ideal Bayesian change-point detection model in which the observer marginalizes over both the current run length (i.e., time since last change) and the current category probability. We compared this model to various alternative models that correspond to different strategies—from approximately Bayesian to simple heuristics—that the observers may have adopted to update their beliefs about probabilities. While a number of models provided decent fits to the data, model comparison favored a model in which probability is estimated following an exponential averaging model with a bias towards equal priors, consistent with a conservative bias, and a flexible variant of the Bayesian change-point detection model with incorrect beliefs. We interpret the former as a simpler, more biologically plausible explanation suggesting that the mechanism underlying change of decision criterion is a combination of on-line estimation of prior probability and a stable, long-term equal-probability prior, thus operating at two very different timescales.

show abstract

Section: Introductionmentioning

confidence: 99%

Human online adaptation to changes in prior probability

Norton

Acerbi

et al. 2019

PLoS Comput Biol

View full text Add to dashboard Cite

show abstract

“…They also found that performance was worse with conditional feedback than with full feedback. Bohil and Wismer (2015) found that implicit learning mediates the influence of base rates on category knowledge acquired through direct experience with feedback on each and every trial. Griffiths and Newell (2009) reported that conditional feedback may actually facilitate learning and offered an alternative to Elwin et al's (2007) constructivist coding hypothesis: 'Selective feedback allows participants to populate their working memory solely with examples which are easy to reason about (high cue-value instances), and this may lead to more rapid learning of the cue-weights and hence more accurate profit predictions' (p. 2889).…”

Section: Effects Of Outcome Feedbackmentioning

confidence: 93%

Detection and Selection Decisions with Conditional Feedback: Interaction of Task Uncertainty and Base Rate

Holzworth

Stewart

Mumpower

2018

Behavioral Decision Making

View full text Add to dashboard Cite

Effects of task uncertainty on decision thresholds in a multiple cue decision task were examined under two types of feedback and three base rate conditions. In most such decision experiments, participants receive feedback after every trial (full feedback) with a single (usually .5) base rate. Our experiment explored conditional (decision-contingent) feedback, in a task representing a detection problem (passenger screening) in which the decision maker receives no feedback unless the decision is positive (search the passenger). Increased uncertainty made all dependent measures worse. Task uncertainty had detrimental effects on both judgment and decision making, and interacted with effects of feedback and base rate. Decision performance was better with full feedback than with conditional feedback, but not by much. There may be no single unifying explanation for results of our base rate manipulation. Conditional feedback generally resulted in fewer positive decisions than full feedback, but not in the low (.1) base rate condition. Results provide partial support for constructivist encoding and for accuracy maximization with moderate and high base rates, but not with a low base rate. Our results indicated lower selection rates with conditional feedback compared with full feedback in moderate and high base rate conditions, and a more exploratory strategy with higher selection rates with conditional feedback compared with full feedback when base rate was low.

show abstract

“…As in prior studies, all participants began by completing an unbiased base-rate “baseline” training phase to separate category structure learning from the effect of base-rates on decision criterion placement [11, 17, 33]. …”

Section: Methodsmentioning

confidence: 99%

“…Bohil & Wismer [ 17 ] recently found support for the hypothesis that implicit learning is important for developing category base-rate sensitivity. They replicated an earlier classification experiment that disrupted implicit learning by use of an observational category training task (participants merely watched stimuli and category labels on training trials as opposed to the more standard design where each stimulus presentation is followed by a response and then feedback about category membership; following [ 25 , 26 ]).…”

Section: Introductionmentioning

confidence: 99%

“…Ashby and colleagues [ 25 ] showed that learning of an easily verbalized category rule—thought to be learned explicitly via conscious, trial-by-trial hypothesis testing—was not diminished by observational training, but learning of a nonverbalizable rule—thought to rely on implicit learning—was substantially disrupted. In a second experiment, Bohil & Wismer [ 17 ] disrupted implicit learning in a response-training condition by inserting a short delay (2.5 seconds) between response and feedback presentation (similar to a category rule-learning study conducted by Maddox, Ashby, & Bohil, [ 27 ]). Both methods of disrupting implicit learning—observational training and response-training with delayed feedback—led to greater conservatism in decision criterion placement (i.e., less response bias toward the high base-rate category).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Base-rate sensitivity through implicit learning

Wismer

Bohil

2017

PLoS ONE

Self Cite

View full text Add to dashboard Cite

Two experiments assessed the contributions of implicit and explicit learning to base-rate sensitivity. Using a factorial design that included both implicit and explicit learning disruptions, we tested the hypothesis that implicit learning underlies base-rate sensitivity from experience (and that explicit learning contributes comparatively little). Participants learned to classify two categories of simple stimuli (bar graph heights) presented in a 3:1 base-rate ratio. Participants learned either from “observational” training to disrupt implicit learning or “response” training which supports implicit learning. Category label feedback on each trial was followed either immediately or after a 2.5 second delay by onset of a working memory task intended to disrupt explicit reasoning about category membership feedback. Decision criterion values were significantly larger following response training, suggesting that implicit learning underlies base-rate sensitivity. Disrupting explicit processing had no effect on base-rate learning as long as implicit learning was supported. These results suggest base-rate sensitivity develops from experience primarily through implicit learning, consistent with separate learning systems accounts of categorization.

show abstract

Implicit learning mediates base rate acquisition in perceptual categorization

Cited by 13 publications

References 21 publications

Human online adaptation to changes in prior probability

Human online adaptation to changes in prior probability

Detection and Selection Decisions with Conditional Feedback: Interaction of Task Uncertainty and Base Rate

Base-rate sensitivity through implicit learning

Contact Info

Product

Resources

About