The Roles of Similarity in Transfer: Separating Retrievability From Inferential Soundness

Similarity is universally acknowledged to be central in transfer, but recent research suggests that its role is complex. The present research attempts to isolate and compare the determinants of similarity-based access to memory and the determinants of the subjective soundness and similarity of a match. We predicted, based on structure-mapping theory, that subjective soundness would depend on the degree of shared relational structure, particularly higher-order structure such as causal bindings. In contrast, we predicted that memory retrieval would be highly sensitive to surface similarities such as common object attributes. To assess retrievability, in three studies, subjects were asked to read a large set of stories and were later given a set of probe stories that resembled the original stories in systematically different ways; e.g., purely relational analogies, surface-similarity matches, or overall (literal similarity) matches. Subjects were told to write out any of the original stories that came to mind. To assess subjective soundness, independent subjects (and also the same reminding subjects) were asked to rate the inferential soundness of each pair; i.e., how well inferences true of one story would apply to the other. As predicted, subjective soundness was highly related to the degree of common relational structure, while retrievability was chiefly related to the degree of surface similarity. Ratings of the similarity of the pairs did not predict the retrievability ordering, arguing against the possibility that the retrieval ordering simply reflected overall similarity. Further, a fourth study demonstrated that subjects given a forced-choice recognition task could discriminate between possible matches on the basis of relational structure, ruling out the possibility that the poor relational retrieval resulted from forgetting or failing to encode the relational structure. We conclude that there is a dissociation between the similarity that governs access to long-term memory and that which is used in evaluating and reasoning from a present match. We describe a model, called MAC/FAC ("any are called but few are chosen"), that uses a two-stage similarity retrieval process to model these findings. Finally, we speculate on the implications of this view for learning and transfer.     

STRUCTURAL ALIGNMENT IN COMPARISON:

Abstract— Theories of similarity generally agree that the similarity of a pair increases with its commonalities and decreases with its differences. Recent research suggests that this comparison process involves an alignment of structured representations yielding commonalities, differences related to the commonalities, and differences unrelated to the commonalities. One counterintuitive prediction of this view is that it should be easier to find the differences between pairs of similar items than to find the differences between pairs of dissimilar items. This prediction is particularly strong for differences that are related to the commonalities. We tested this prediction in two experiments in which subjects listed a single difference for each of a number of word pairs. The results are consistent with the predictions of structural alignment. In light of these findings, we discuss the potential role of structural alignment in other cognitive processes that involve comparisons.     

A temporal ratio model of memory

A model of memory retrieval is described. The model embodies 4 main claims: (a) temporal memory--traces of items are represented in memory partly in terms of their temporal distance from the present; (b) scale-similarity--similar mechanisms govern retrieval from memory over many different timescales; (c) local distinctiveness--performance on a range of memory tasks is determined by interference from near psychological neighbors; and (d) interference-based forgetting--all memory loss is due to interference and not trace decay. The model is applied to data on free recall and serial recall. The account emphasizes qualitative similarity in the retrieval principles involved in memory performance at all timescales, contrary to models that emphasize distinctions between short-term and long-term memory.     

Are the inhibitory and faciliatory effects during retrieval of semantically related items present in amnestic mild cognitive impairment?

Prefrontal functions subserve inhibition control for retrieval of semantically related items inducing forgetting 19 a-MCI patients and 29 controls underwent neuropsychological evaluation and retrieval-practice paradigm (RPP) to estimate baseline remember (BR), retrieval-induced facilitation (FAC) and retrieval-induced forgetting (RIF). A-MCI patients underwent also 3 T-MRI to assess relationship between regional grey matter (rGM) volumes and RPP indexes Behaviourally, RIF and FAC were both observed controls, while RIF only was observed in a-MCI patients. In patients but not in controls, RIF was associated with cognitive efficiency and FAC with memory performance. Patients showed also associations between BR and rGM volumes in the precuneus, no association was found between rGM volumes and RIF and FAC. A-MCI patients did not benefit from repeated practice during retrieval of studied items, which is likely due to their memory disorder. In contrast, patient cognitive efficiency would drive retrieval suppression of interfering stimuli.     

An attractor network model of serial recall

We present a neural network model of verbal working memory which attempts to illustrate how a few simple assumptions about neural computation can shed light on cognitive phenomena associated with the serial recall of verbal material. We assume that neural representations are distributed, that neural connectivity is massively recurrent, and that synaptic efficacy is modified based on the correlation between pre- and post-synaptic activity (Hebbian learning). Together these assumptions give rise to emergent computational properties that are relevant to working memory, including short-term maintenance of information, time-based decay, and similarity-based interference. We instantiate these principles in a specific model of serial recall and show how it can both simulate and explain a number of standard cognitive phenomena associated with the task, including the effects of serial position, word length, articulatory suppression (and its interaction with word length), and phonological similarity.     

单字法有意遗忘的内部心理机制：基于多项式加工树模型的分析

通过构建使用单个双字词作为实验材料,并在测验阶段全面考察记忆提取过程的MPT模型,考察单字法有意遗忘的心理机制。结果发现,对于TBF/R项目,遗忘组的储存能力和提取能力均弱于记忆组;对于TBR/R项目,遗忘组的储存能力强于记忆组。并且无论对于短时SOA还是长时SOA,遗忘组对TBF项目后探测刺激的反应时都显著长于TBR项目。研究表明,单字法有意遗忘现象是由选择性复述、注意抑制和提取抑制三种机制共同作用的结果。     

A feature model of immediate memory

A feature model of immediate memory is presented, and simulations are described. List items are characterized as multiattribute vectors that can be selectively overwritten by subsequent external events and by the ongoing stream of internal activity. Degraded primary memory vectors are compared with intact secondary memory vectors, and retrieval likelihood is computed as the ratio of similarities. The model is shown to account for the major modality-based phenomena of the immediate serial recall literature, including modality-based temporal grouping effects and the negative effects of phonological similarity.     

Extending SME to Handle Large‐Scale Cognitive Modeling

Analogy and similarity are central phenomena in human cognition, involved in processes ranging from visual perception to conceptual change. To capture this centrality requires that a model of comparison must be able to integrate with other processes and handle the size and complexity of the representations required by the tasks being modeled. This paper describes extensions to Structure‐Mapping Engine (SME) since its inception in 1986 that have increased its scope of operation. We first review the basic SME algorithm, describe psychological evidence for SME as a process model, and summarize its role in simulating similarity‐based retrieval and generalization. Then we describe five techniques now incorporated into the SME that have enabled it to tackle large‐scale modeling tasks: (a) Greedy merging rapidly constructs one or more best interpretations of a match in polynomial time: O(n²log(n)); (b) Incremental operation enables mappings to be extended as new information is retrieved or derived about the base or target, to model situations where information in a task is updated over time; (c) Ubiquitous predicates model the varying degrees to which items may suggest alignment; (d) Structural evaluation of analogical inferences models aspects of plausibility judgments; (e) Match filters enable large‐scale task models to communicate constraints to SME to influence the mapping process. We illustrate via examples from published studies how these enable it to capture a broader range of psychological phenomena than before.     

Do development and learning really decrease memory? On similarity and category-based induction in adults and children

In a recently published study, Sloutsky and Fisher [Sloutsky, V. M., & Fisher, A.V. (2004a). When development and learning decrease memory: Evidence against category-based induction in children. Psychological Science, 15, 553-558; Sloutsky, V. M., & Fisher, A. V. (2004b). Induction and categorization in young children: A similarity-based model. Journal of Experimental Psychology: General, 133, 166-188.] demonstrated that children have better memory for the items that they generalise to than do adults. On the basis of this finding, they claim that children and adults use different mechanisms for inductive generalisations; whereas adults focus on shared category membership, children project properties on the basis of perceptual similarity. Sloutsky & Fisher attribute children's enhanced recognition memory to the more detailed processing required by this similarity-based mechanism. In Experiment 1 we show that children look at the stimulus items for longer than adults. In Experiment 2 we demonstrate that although when given just 250ms to inspect the items children remain capable of making accurate inferences, their subsequent memory for those items decreases significantly. These findings suggest that there are no necessary conclusions to be drawn from Sloutsky & Fisher's results about developmental differences in generalisation strategy.     

Putting Short-Term Memory Into Context: Reply to Usher, Davelaar, Haarmann, and Goshen-Gottstein (2008)

The temporal context model posits that search through episodic memory is driven by associations between the multiattribute representations of items and context. Context, in turn, is a recency weighted sum of previous experiences or memories. Because recently processed items are most similar to the current representation of context, M. Usher, E. J. Davelaar, H. J. Haarmann, and Y. Goshen-Gottstein (2008) have suggested that the temporal context model (TCM-A) embodies a distinction between short-term and long-term memory and that this distinction is central to TCM-A's success in accounting for the pattern of recency and contiguity observed across short and long timescales. The authors dispute Usher et al's claim that context in TCM-A has much in common with classic interpretations of short-term memory. The idea that multiple representations interact in the process of memory encoding and retrieval (across timescales), as proposed in TCM-A, is very different from the classic dual-process view that distinct rules govern retrieval of recent and remote memories.     

Integrating structure and meaning: a distributed model of analogical mapping

In this paper we present Drama , a distributed model of analogical mapping that integrates semantic and structural constraints on constructing analogies. Specifically, Drama uses holographic reduced representations (Plate, 1994), a distributed representation scheme, to model the effects of structure and meaning on human performance of analogical mapping. Drama is compared to three symbolic models of analogy (SME, Copycat, and ACME) and one partially distributed model (LISA). We describe Drama's performance on a number of example analogies and assess the model in terms of neurological and psychological plausibility. We argue that Drama's successes are due largely to integrating structural and semantic constraints throughout the mapping process. We also claim that Drama is an existence proof of using distributed representations to model high-level cognitive phenomena.     

The phonological similarity effect in immediate recall: Positions of shared phonemes

Earlier literature proposes two ways phonological similarity could harm immediate recall: (1) It could increase the degradation of the representations of items in memory, or (2) it could decrease the probability that a degraded representation is correctly reconstructed. A multinomial processing tree model for each hypothesis was used to analyze an immediate recall experiment. Both gave a good account of the data, but, of the two, results favor the hypothesis that the effect of phonological similarity is to impair reconstruction of degraded representations. A second issue is whether positions of repeated phonemes in phonologically similar items matter. We found that mere repetition of phonemes produced a phonological similarity effect. Repeated phonemes in the same positions appeared to produce a greater effect. A final finding is that when reading rate was preequated, phonological similarity affected memory span by changing the time taken to recall a list of span length.     

Similarity-based interference in a working memory numerical updating task: age-related differences between younger and older adults

Similarity among representations held simultaneously in working memory (WM) is a factor which increases interference and hinders performance. The aim of the current study was to investigate age-related differences between younger and older adults in a working memory numerical updating task, in which the similarity between information held in WM was manipulated. Results showed a higher susceptibility of older adults to similarity-based interference when accuracy, and not response times, was considered. It was concluded that older adults' WM difficulties appear to be due to the availability of stored information, which, in turn, might be related to the ability to generate distinctive representations and to the process of binding such representations to their context when similar information has to be processed in WM.     

Working memory retrieval differences between medial temporal lobe epilepsy patients and controls: A three memory layer approach

Multi-store models of working memory (WM) have given way to more dynamic approaches that conceive WM as an activated subset of long-term memory (LTM). The resulting framework considers that memory representations are governed by a hierarchy of accessibility. The activated part of LTM holds representations in a heightened state of activation, some of which can reach a state of immediate accessibility according to task demands. Recent neuroimaging studies have studied the neural basis of retrieval information with different states of accessibility. It was found that the medial temporal lobe (MTL) was involved in retrieving information within immediate access store and outside this privileged zone. In the current study we further explored the contribution of MTL to WM retrieval by analyzing the consequences of MTL damage to this process considering the state of accessibility of memory representations. The performance of a group of epilepsy patients with left hippocampal sclerosis in a 12-item recognition task was compared with that of a healthy control group. We adopted an embedded model of WM that distinguishes three components: the activated LTM, the region of direct access, and a single-item focus of attention. Groups did not differ when retrieving information from single-item focus, but patients were less accurate retrieving information outside focal attention, either items from LTM or items expected to be in the WM range. Analyses focused on items held in the direct access buffer showed that consequences of MTL damage were modulated by the level of accessibility of memory representations, producing a reduced capacity.     

The dual effect of context on memory of related and unrelated themes: Discrimination at encoding and cue at retrieval

The influence of contextual factors on encoding and retrieval in recognition memory was investigated using a retroactive interference paradigm. Participants were randomly assigned to four context conditions constructed by manipulating types of presentation modality (pictures vs words) for study, interference, and test stages, respectively (ABA, ABB, AAA, & AAB). In Experiment 1 we presented unrelated items in the study and interference stages, while in Experiment 2 each stage contained items from the same semantic category. The results demonstrate a dual role for context in memory processes-at encoding as well as at retrieval. In Experiment 1 there is a hierarchical order between the four context conditions, depending on both target-test and target-interference contextual similarity. Adding a categorical context in Experiment 2 helped to specify each list and therefore better distinguish between target and interferer information, and in some conditions compensated for their perceptual similarity.     

Similarity-based cognition: radical enactivism meets cognitive neuroscience

Similarity-based cognition is commonplace. It occurs whenever an agent or system exploits the similarities that hold between two or more items—e.g., events, processes, objects, and so on—in order to perform some cognitive task. This kind of cognition is of special interest to cognitive neuroscientists. This paper explicates how similarity-based cognition can be understood through the lens of radical enactivism and why doing so has advantages over its representationalist rival, which posits the existence of structural representations or S-representations. Specifically, it is argued that there are problems both with accounting for the content of S-representations and with understanding how neurally-based structural similarities can work as representations (even if contentless) in guiding intelligent behavior. Finally, with these clarifications in place, it is revealed how radical enactivism can commit to an account of similarity-based cognition in its understanding of neurodynamics.

     

Using context to build semantics

Latent semantic analysis (LSA) is a model of knowledge representation for words. It works by applying dimension reduction to local co-occurrence data from a large collection of documents after performing singular value decomposition on it. When the reduction is applied, the system forms condensed representations for the words that incorporate higher order associations. The higher order associations are primarily responsible for any semantic similarity between words in LSA. In this article, a memory model is described that creates semantic representations for words that are similar in form to those created by LSA. However, instead of applying dimension reduction, the model builds the representations by using a retrieval mechanism from a well-known account of episodic memory.     

Induction and categorization in young children: a similarity-based model

The authors present a similarity-based model of induction and categorization in young children (SINC). The model suggests that (a). linguistic labels contribute to the perceived similarity of compared entities and (b). categorization and induction are a function of similarity computed over perceptual information and linguistic labels. The model also predicts young children's similarity judgment, induction, and categorization performance under different stimuli and task conditions. Predictions of the model were tested and confirmed in 6 experiments, in which 4- to 5-year-olds performed similarity judgment, induction, and categorization tasks using artificial and real labels (Experiments 1-4) and recognition memory tasks (Experiments 5A and 5B). Results corroborate the similarity-based account of young children's induction and categorization, and they support both qualitative and quantitative predictions of the model.     

MINERVA 2: A simulation model of human memory

An overview of a simulation model of human memory is presented. The model assumes: (1) that only episodic traces are stored in memory, (2) that repetition produces multiple traces of an item, (3) that a retrieval cue contacts all memory traces simultaneously, (4) that each trace is activated according to its similarity to the retrieval cue, and (5) that all traces respond in parallel, the retrieved information reflecting their summed output. The model has been applied with success to a variety of phenomena found with human subjects in frequency and recognition judgment tasks, the schema-abstraction task, and paired-associate learning. Application of the model to these tasks is briefly summarized.     

Processes of similarity judgment

Similarity underlies fundamental cognitive capabilities such as memory, categorization, decision making, problem solving, and reasoning. Although recent approaches to similarity appreciate the structure of mental representations, they differ in the processes posited to operate over these representations. We present an experiment that differentiates among extant structural accounts of similarity in their ability to account for patterns of similarity ratings. These data pose a challenge for transformation-based models and all but one mapping-based model, the Similarity as Interactive Activation and Mapping (SIAM) model of similarity.