A memory-based scaling model—ANCHOR—is proposed and tested. The perceived magnitude of the target stimulus is compared with a set of anchors in memory. Anchor selection is probabilistic and sensitive to similarity, base-level strength, and recency. The winning anchor provides a reference point near the target and thereby converts the global scaling problem into a local comparison. An explicit correction strategy determines the final response. Two incremental learning mechanisms update the locations and base-level activations of the anchors. This gives rise to sequential, context, transfer, practice, and other dynamic effects. The scale unfolds as an adaptive map. A hierarchy of models is tested on a battery of quantitative measures from 2 experiments in absolute identification and category rating. Category rating is a widely used method of data collection in experimental psychology. Ratings come in a wide variety of guises: psychophysical scales, similarity judgments, typicality judgments, confidence ratings, attitude questionnaires, health selfreports, and many others. The participants in all these tasks are asked to rate things using an ordered set of categories such as 1,..., 7 or strongly agree,..., strongly disagree. Most people

This paper illustrates how mechanisms initially designed for explaining analogy-making can also model judment and choice and account for contextual effects on these processes. The JUDGEMAP model is presented as well as three simulations that replicate some well known contextual effects in judgment and choice. It is demonstrated how the same basic mechanisms, designed for analogy-making can be responsible for seemingly unrelated phenomena, like the frequency effect in judgment and the concave form of the utility function; the preference for the middle ratings in judgment, the nonlinear form of subjective probability; and the effect of preference reversal.

Abstract. This paper is testing a DUAL-based model of memory. The model assumes decentralized representation of episodes as a coalition of agents and analogical transfer processes as the basis for memory reconstruction of our past. It is a model of active reconstruction thereby allowing memory insertions and blending of episodes. The experiment explores the role of the degree of internal connectivity of the coalition representing the episode on the outcome of the reconstruction process. It demonstrates that the more the links between the elements of the episode are, the higher the number of details we recall, and the lesser the intruded elements and the context influence.

The paper presents an experiment that tests the influence of an irrelevant to the task feature (namely the color of the fonts used) on the judgment of rent prices of various apartments. The rents were presented with either green or red digits that stand for the price of the apartments. Participants judged how expensive the prices are from their perspective on a 7 point scale. As result it turned out that the rents obtained different ratings depending on the color with which they were presented, i.e. there was an influence of the irrelevant to the task dimension “color ” on human judgment of a quite abstract dimension such as price. The difference is small but significant. Three possible explanations of these contextual effects are discussed with respect to the specific material used in this experiment. It was argued that spreading activation (Kokinov et al., 2004, Petkov, 2005a, 2005b) provides a better explanation of the reported results than the low-level

This paper presents two experiments exploring context effects on human judgment and testing JUDGEMAP’s predictions that irrelevant information such as the color of the stimulus can produce a contrast effect with respect to skewed stimuli set to be judged. The first experiment demonstrates the effect on a perceptual judgment task (judging the length of lines which are colored in red and green), while the second experiment demonstrates exactly the same effect on a conceptual judgment task (judging the age of a person when the absolute age is presented by numbers with colored digits – red and green). These results rule out the “recalibration ” of the perceptual system explanation. The fact that the same effects were obtained in both cases prompts for a common explanation. JUDGEMAP provides such an explanation: the color is used in retrieving past instances in WM based on the spreading activation mechanism and thus the comparison set becomes biased. The biased comparison set which is then mapped on the scale elements produces a shift in the mean ratings.

Why do people judge green lines to be shorter/longer than red lines of the same length? Do they perceive them differently or do they retrieve a biased set of alternatives in their comparison set?