Retrospective Revaluation as Simple Associative Learning.

Ghirlanda, Stefano

doi:10.1037/0097-7403.31.1.107

Cited by 31 publications

(29 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Minerva-AL's explanation of retrospective revaluation as de novo learning via within-compound associations is consistent with a number of other models (e.g., Dickinson & Burke, 1996;Ghirlanda, 2005;Miller & Matzel, 1988;Van Hamme & Wasserman, 1994;Witnauer & Miller, 2011). However, Minerva-AL offers a unique computational description of how that process unfolds by a combination of discrepancy encoding and trace inversion at retrieval.…”

Section: Section 5: Retrospective Revaluationmentioning

confidence: 75%

An instance theory of associative learning

2011

View full text Add to dashboard Cite

We present and test an instance model of associative learning. The model, Minerva-AL, treats associative learning as cued recall. Memory preserves the events of individual trials in separate traces. A probe presented to memory contacts all traces in parallel and retrieves a weighted sum of the traces, a structure called the echo. Learning of a cue-outcome relationship is measured by the cue's ability to retrieve a target outcome. The theory predicts a number of associative learning phenomena, including acquisition, extinction, reacquisition, conditioned inhibition, external inhibition, latent inhibition, discrimination, generalization, blocking, overshadowing, overexpectation, superconditioning, recovery from blocking, recovery from overshadowing, recovery from overexpectation, backward blocking, backward conditioned inhibition, and second-order retrospective revaluation. We argue that associative learning is consistent with an instance-based approach to learning and memory.Keywords Associative learning . Memory . Instance theory . Exemplar theoryIn a simple associative learning procedure, a cue, A, is presented followed by an outcome, X. With experience, presentation of A elicits anticipation of X. The growth of anticipation is the process of associative learning.Typically, associative learning is modeled as a gradual accrual of excitatory and inhibitory connections between stimulus units. Associative strength is treated as summative; so the strength of an association between stimuli on trial t stands for the entire history of learning. This scheme for understanding learning is well described in the RescorlaWagner model (Rescorla & Wagner, 1972) and the many theories that follow from it.Whereas summative theories provide insight into learning, they make an unreasonable assumption. Namely, they deny that the learner remembers the events of separate learning trials-a problem that Miller, Barnet, and Grahame (1995) call the assumption of path independence. The problem is important. First, it distinguishes learning from memory. Second, it denies a growing body of evidence that animals other than humans remember the events of learning trials (e.g., Fagot & Cook, 2006;Vaughan & Greene, 1984;Voss, 2009).In contrast to the classical theories of learning, instance theories of human memory identify the individual experience (i.e., the instance) as the primitive unit of knowledge and treat learning as the accumulation and deployment of instances from memory. Brooks (1978Brooks ( , 1987 was among the first to champion the view. Medin and Schaffer (1978) were among the first to formalize it (see also Reed, 1972). Hintzman's (1986Hintzman's ( , 1988 Minerva 2 model and Nosofsky's (1986) generalized context model are classic formalizations of the view. Kruschke's (1992Kruschke's ( , 1996Kruschke's ( , 2001) and Logan's (1988Logan's ( , 2002 Learn Behav (2012) 40:61-82 DOI 10.3758/s13420-011-0046-2 In this article, we build on previous efforts (Jamieson, Hannah & Crump 2010) to show that an instance model of ass...

show abstract

Section: Section 5: Retrospective Revaluationmentioning

confidence: 75%

An instance theory of associative learning

2011

View full text Add to dashboard Cite

show abstract

“…The Rescorla-Wagner model handily accounts for blocking, but because it assumes absent cues have zero influence on learning, it cannot account for backward blocking. Extensions of the model, that assume absent cues have a negative impact on learning, can account for backward blocking or other effects (Dickinson & Burke, 1996;Ghirlanda, 2005;Markman, 1989;Tassoni, 1995;Van Hamme & Wasserman, 1994). The models assert that only absent cues that are expected to be present should have negative impact, but the exact computations regarding which cues are expected, and their magnitude of negativity, have been left unspecified.…”

Section: Application To Blocking and Backward Blockingmentioning

confidence: 99%

Locally Bayesian learning with applications to retrospective revaluation and highlighting.

Kruschke¹

2006

Psychological Review

108

View full text Add to dashboard Cite

A scheme is described for locally Bayesian parameter updating in models structured as successions of component functions. The essential idea is to back-propagate the target data to interior modules, such that an interior component's target is the input to the next component that maximizes the probability of the next component's target. Each layer then does locally Bayesian learning. The approach assumes online trial-by-trial learning. The resulting parameter updating is not globally Bayesian but can better capture human behavior. The approach is implemented for an associative learning model that first maps inputs to attentionally filtered inputs and then maps attentionally filtered inputs to outputs. The Bayesian updating allows the associative model to exhibit retrospective revaluation effects such as backward blocking and unovershadowing, which have been challenging for associative learning models. The back-propagation of target values to attention allows the model to show trial-order effects, including highlighting and differences in magnitude of forward and backward blocking, which have been challenging for Bayesian learning models.

show abstract

“…Furthermore, the importance or strength of correct (β 1 ) vs. incorrect (β 2 ) predictions can be weighted differentially. Although in a typical simulation run these parameters are set to their default values, they allow for a principled account of various learning ÒpeculiaritiesÓ (see Ghirlanda, 2005 for how a simple error-driven learning model, formally equivalent to the Rescorla-Wagner model, can account for a range of intricate learning phenomena). For example, in a series of experiments and modelling studies it was shown that a change in the salience of a crucial learning cue could explain an unexpected pattern of results in a variant of the object naming task, designed to demonstrate the highlighting effect in learning.…”

Section: Learning Theorymentioning

confidence: 99%

Towards cognitively plausible data science in language research

Milin

Divjak

Dimitrijević

et al. 2016

Cognitive Linguistics

View full text Add to dashboard Cite

Over the past 10 years, Cognitive Linguistics has taken a Quantitative Turn. Yet, concerns have been raised that this preoccupation with quantification and modelling may not bring us any closer to understanding how language works. We show that this objection is unfounded, especially if we rely on modelling techniques based on biologically and psychologically plausible learning algorithms. These make it possible to take a quantitative approach, while generating and testing specific hypotheses that will advance our understanding of how knowledge of language emerges from exposure to usage.

show abstract

Retrospective Revaluation as Simple Associative Learning.

Cited by 31 publications

References 33 publications

An instance theory of associative learning

An instance theory of associative learning

Locally Bayesian learning with applications to retrospective revaluation and highlighting.

Towards cognitively plausible data science in language research

Contact Info

Product

Resources

About